At this point, a voluntary “2 to 4 million acre feet of additional conservation” Colorado River deal by Aug. 16 seems out of reach

Janet Wilson had a helpful story yesterday in the Desert Sun about California’s negotiations over its piece of the looming Colorado River cutbacks. Its bottom line is that California – the state with the largest Colorado River allocation – is talking about kicking in 500,000 acre feet of water. Or maybe it’s really just 400,000 acre feet of water – as the Metropolitan Water District of Southern California’s Bill Hasencamp told her, paraphrased, the negotiations are fluid and numbers could change.

A reminder of what Reclamation Commissioner Camille Touton told senators just seven weeks ago:

In the Colorado River Basin more conservation and demand management are needed in addition to the actions already underway. Between 2 and 4 million acre feet of additional conservation is needed just to protect critical elevations in 2023. (emphasis added)

4 million acre feet is obviously out of reach. It always was.

But if Wilson’s numbers about California’s contributions are right – and she’s a good reporter, we have every reason to believe they’re in the ballpark – 2 million acre feet of additional conservation is beyond the grasp of a voluntary deal as well.

The arithmetic is straightforward.

The Upper Basin has said “not our problem“.

Nevada’s share of the river is so tiny that its contribution is couch cushion change, a rounding error.

That leaves, in round numbers, 1.5 million acre feet of water to come out of Arizona just to get to Touton’s bottom line number for additional conservation. That would require completely drying up the Central Arizona Project canal. (CAP is taking 1.031maf this year, and averaged ~1.4maf over the previous fives years). I’m frequently surprised by Arizona, but it seems unlikely that they’ll agree to a voluntary deal that dries up the CAP canal. If that’s where we end up, Arizona’s best alternative to a negotiated agreement will be to just make the feds do it, make them take the heat. (Worth noting that FiveThirtyEight has Arizona Democrat Mark Kelly slightly favored to hold his seat. Water politics is high stakes politics.)

Combine that with the reality that Arizona’s Native American communities, major water rights holders, have complained that they’ve been cut out of this entire process, according to a July 22 letter just surfaced.

I can imagine creative accounting that might allow everyone to grin through their teeth and count water moved down to Lake Powell from Flaming Gorge and other Upper Basin reservoirs as part of the 2 million. That’s pretty clearly not what Touton called for in June. It’s not “additional conservation”. But it might create some space for a face-saving deal.

Whether that would be enough to protect us from dead pool is another question.

A reminder of the stakes

The Bureau of Reclamation’s most recent “minimum probable” model runs show Lake Powell dropping below power pool – unable to generate electricity, and forced to move water through bypass tubes that Reclamation has made clear it does not trust – by October 2023.

Under that same scenario, Lake Mead drops to elevation 992 feet above sea level over the next 24 months.

(Trust me, having to type a Lake Mead elevation level without having to use a comma made me clench.)

At that point, a lack of water will make massive cuts a self-executing reality. We’ve drained our buffer. You can’t use water that doesn’t exist.



  1. Colorado, as the first diverter of water out of the Colorado river basin, needs to realize that they are almost as big a user as Los Angeles (Metropolitan Water District). Time for Colorado to do their part.
    All of the Colorado front range water, all of the green lawns in Denver, all of the water going to the Platte River and into Nebraska, needs to be in the conversation.
    The avg water use per capita is a number that would be fair to determine where to cut water. Those areas or states still using large amounts of water per capita should be the first and largest conservation contributors.

  2. The reason this is so hard is that it is ALL about growth.

    We would have enough rain in Tucson (max potential is 80B gal/yr) for 1M people, including as much of our food as we choose to grow.

    What we don’t have is enough water to support unending growth. It’s all about keeping the growth machine going for as long as possible.

  3. Thanks for the update, John.

    What do you mean when you write that Reclamation doesn’t trust the bypass tubes? Doesn’t fear they will fail structurally or something?

  4. Le Chatlier rules. No one has ever reassemble Humpty Dumpty after he fell. Time to enforce the IID DCP and private water by eminent domain and pay for it. With ag water gone we can now starve and become food dependencies of Chile and Mexico. That will surely mobilize Congress to enact an American Food act. Substitute food for chips. Sell your Lower Basin real estate and move north to Canada which has 30% of the worlds fresh water and only 30 million inhabitants. Bright future. Yellowknife, here I come.

  5. 4 million AF cuts are not enough. The flow from snow melt this water year was 3.6MAF. The usage for the system is 14 MAF. We have a 10 MAF mismatch. Future flows are being modeled at 11MAF/year. If that is correct the flows the next two years have to come in at 17 MAF to average 11MAF for water years 2021-2024. Who believes that?

    Until the parties involved get a handle on this simple arithmetic things will keep getting worse. This is a fixable problem but only if you know how to add numbers together and have the courage to make unpopular decisions.

    We have to use considerably less than the natural stream flow if we are to rebuild the reservoir storage for the system to function in the long term.

  6. one added point:

    Nevada’s share is arguably by no means a rounding error. Nevada’s consumptive use is 300,000 acre-feet per year, that is 8% of the water available from snowmelt this water year (April-July 2022 Powell inflows). Data from the CBRFC and SNWA.

  7. The minimal percentage of Nevada’s usage of the Colorado River is irrelevant to Nevadans, because Nevada has such a small population. What is relevant is how it will affect Nevada directly? What percentage of Nevada’s needs are affected? That is the pertinent question. The 2% (or 8%) number that people bandy around may be true, but it is not relevant to the Nevada problem. What is more critical is, when Lake Mead becomes a dead pool, what percentage of water will Las Vegas, and Boulder City, will be lost? What will be left? That’s what matters.

  8. Basin-wide, about 70-80% of water use is for agriculture. So most Colorado River reduction will come from agriculture, especially given the economic value of different water uses. Cities will reduce the cost of them acquiring agricultural water through additional water conservation. City per capita water use usually still can slim down quite a bit.

    The point that substantial additional use reduction will be needed to to somewhat recover reservoir levels is a very good one, unless the next few years are very hydrologically lucky.

  9. @ George Rhee

    You’re comparing apples and oranges here. The 3.6 MAF number for 2022 “flow from snowmelt” you cite is the *unregulated* inflow to Powell (and for April-July, not the whole water year). The unregulated inflow already reflects the ~4 MAF of Upper Basin consumptive use that occurs above Powell.

    The other flow numbers you cite (11 MAF; 17 MAF) are *natural* flows at Lees Ferry for the water year–to calculate these, UB consumptive use is added back in.

    For Water Year 2022, the Lees Ferry natural flow looks like it will come in around 9.5 MAF…which is not good, but a far cry from 3.6 MAF.

  10. This conversation is likely to change dramatically once we see what, or if there are, any changes enacted by the lower basin Water Master. The “sword of Damocles” situation required cut back agreement will not encourage a compromise unless some type of decision is handed down In August, or at least this fall. It will be an interesting next couple of months on the Colorado River…

  11. To Jeff Lukas point. I specifically asked Cody Moser at CBRFC this question about the definition of unregulated inflow. He said unregulated inflow is essentially the natural streamflow of the river.

    From the CBRFC definition on their website:
    “Unregulated inflow: Flow (or volumetric water supply) where known, measured diversions, imports,
    and reservoir regulation are accounted for to approximate natural flow conditions of the river.”

    I asked about the discrepancy between the Lees Ferry natural flows from USBR and the CBRFC numbers and he suggested to talk to USBR. I agree these differences do need to be reconciled. Part of the river problems arise from people consistently taking the more optimistic numbers to assess the situation and consequently underestimating the magnitude of the risk exposure, leading inevitably to the current emergency.

  12. The coming conversations should all be about; What do we want to use the river flow for?
    Population growth?
    Food growth?
    Environmental improvement?
    Meet the Mexican requirements?
    Once we have that figured out, then dividing the water will be easier.

    Since water rights rules were tossed out in order to create the 1922 Compact, not sure that current water rights will be a part of the coming decisions.

  13. @George Rhee

    You raise an important issue and a point of frequent confusion: CBRFC and Reclamation have different operational definitions of “unregulated flows.”

    CBRFC: Adjusted for measured diversions and imports, and upstream reservoir regulation (but not unmeasured diversions or unmeasured return flows)

    Reclamation: Adjusted only for upstream reservoir regulation

    Cody Moser is correct in that for most of CBRFC’s 150+ forecast points, their unregulated flows are close to fully naturalized flows.

    However, for three of CBRFC’s forecast points–inflows to Flaming Gorge Res., Navajo Res., and Powell–CBRFC effectively defers to the Reclamation definition, adjusting their forecasted unregulated volumes for those three points via linear regression to match the more limited naturalization that Reclamation uses (i.e., only adjusting for upstream reservoir regulation).

    We did a comparison for the 2020 State of the Science report, and found that CBRFC’s published April-July unregulated Powell inflows from 1964-2016 ( agree almost perfectly with Reclamation’s published unregulated inflows, with differences averaging 0.02%.

    In any case: Whether you get the numbers from CBRFC or Reclamation, the “unregulated inflows” for Lake Powell are not fully naturalized; most significantly, they are not adjusted for Upper Basin consumptive use, and so they will be lower than Reclamation’s natural flows at Lees Ferry (which are adjusted for UB CU), by roughly ~4 MAF on a water-year basis.

  14. Wiping out the economy of 40 million people – whose share of the Nation’s economy is 1/5th to1/6. One commentor says that Wall Street would collapse if the reservoir system was drained. Clearly large impact and shock waves throughout the society of the West. If such a blow to the personal economy of 11 percent of the Nation’s people did occur it would be tragic and profound. Now is past the time to act. #adultsintheroom

Comments are closed.