For the latest “Ask me a Colorado River question,” a friend of Inkstain wonders:

Q: Can you move water from one Colorado River Basin state to another?

A: No.

OK, this is the Law of the River, so of course it’s more complicated than that, with numerous edge case exceptions, but basically, no, you can’t move water from one Colorado River Basin state to another.

The 1922 Colorado River Compact divided what the framers thought was a reasonable estimate of the Colorado River’s flow in half. States of the “Upper Basin” – Wyoming, Colorado, Utah and New Mexico – get 7.5 million acre feet per year, and state of the Lower Basin get 7.5 million acre feet per year.

The Upper Basin states subsequently got together to split up their share on a percentage basis:

• Colorado 51.75%
• Utah 23%
• Wyoming 14%
• New Mexico 11.25%

The Lower Basin states fussed and feuded and finally the Supreme Court stepped in and said:

• California 4.4 million acre feet
• Arizona 2.8 million acre feet
• Nevada 300 thousand acre feet

Each state then is left to its own devices to divide up its own share internally, based on its own internal water laws and politics. But one of those devices cannot be to buy some extra water from one of the other states.

The result is that, to the extent that we’re seeing innovative water sharing and transfer agreements, they all tend to happen within a single state.

In Arizona, for example, the Wellton-Mowhawk and Yuma Mesa irrigation districts are involved in deals that involve water conservation and transfer to urban water users. In California, the Palo Verde Irrigation District has a program in which land can be fallowed and saved water shipped off to Los Angeles. The Imperial Irrigation District has an agreement in which agricultural water conservation measures generate water that is then used by urban users in San Diego.

I say they all tend to happen within a single state, though, because there are interesting edge cases where boundaries are being tested. Nevada and Arizona, for example, have a program in which surplus Nevada water is banked in Arizona aquifers for later use. Complicated accounting swaps ensue through which, on paper, no water crosses state lines. Wink wink.

New Mexico is currently considering developing a New Mexico unit of the Central Arizona Project which would involved diverting Gila River water within New Mexico, before it gets to Arizona, and then (again on paper) swapping that water with downstream CAP water.

But the most interesting test of these boundaries is an effort currently under development that Henry Brean wrote about in May under which urban agencies across the basin would pool their resources to fund conservation programs (most likely ag conservation) somewhere in the basin. The saved water would simply stay in the river for everyone’s shared use – “system water”. The result is that water conserved in State A might then be used in State B, but no one’s accounting for it specifically. It’s a clever workaround.

So the answer to the original question is really, “Not exactly, but people are trying.”

Background:

• Overall Law of the River from the U.S. Bureau of Reclamation
• The Arizona-Nevada water banking program
• System conservation program

update: With the latest hourly reading, the elevation of Lake Mead has dropped to 1,081.84 feet above sea level, the lowest it has been since the reservoir was being filled in the 1930s:

With apologies to Emily Green, Lake Mead – Low Bad

Previously: Last night at midnight, the surface elevation of Lake Mead dropped to 1,081.89 feet above sea level, half an inch above the previous record low of 1,081.85 set around 9 p.m. Nov. 27, 2010 (source pdf). There’s a certain ebbing and flowing as water flows in from upstream and flows out to run turbines and water downstream farms, so I can’t say with precision when we’ll break the old mark, but by some time tonight or early tomorrow, it’s likely to drop past the record.

This already generating a lot of attention, and is likely to generate more as the record is set. (I had to wait in line this morning behind a TV station when I wanted to talk to one of the key basin water managers. I am not in a hurry. I am writing a book, which has a looser deadline.)

One of the truisms you will likely hear is that the drought since 2000, which has contributed to Mead’s decline, is unprecedented, outside of the historical record and therefore something water managers could not have been expected to plan for. But I recently ran across an interesting graph from the Bureau of Reclamation that suggests that may not be quite right. If we take the estimated river flows of the drought conditions that spread across the Colorado River Basin in the 1930s and early 1940 and overlay them atop a 21st-century water management system, Lake Mead would actually drop faster than it is today.

Simulated Lake Mead levels, courtesy USBR

This is from the 2007 Final Environmental Impact Statement for the Colorado River Interim Guidelines for Lower Basin Shortages and Coordinated Operations for Lakes Powell and Mead. (source) The hydrologists created a series of synthetic data series and plugged them into a modern river system operational model – dams, water users and all – to see what would happen to Lake Mead in the future under various inflow scenarios.

Look at how that blue “Trace 1” line crashes beginning in 2031. That is a simulation of what would happen to Lake Mead if the drought of the 1930s-1950s was repeated in the 21st century. Under that scenario, it takes Lake Mead 14 years to drop from full to a surface elevation of 1,075. The current best estimates are that we could hit 1,075 in 2015 (source pdf), 15 years after it was last fullish in 2000. But what’s got to be unnerving if you’re a water manager is what happens to Trace 1 after it hits 1075. Under that historical hydrology, Lake Mead kept dropping for nearly a decade.

So what’s happening now is essentially equal to, but perhaps not worse, than, the worst case in the historical record.

Caveat: Here’s where the apples and oranges mix. The “Trace 1” scenario, in addition to taking the historical hydrology, presumes some expansion of water use in the future. So by 2031, when Trace 1 begins its crash, there is perhaps ~250,000 acre feet of additional water use in the Upper Basin that is included in the model.