Resistance is Futile – Agriculture is Key to Fixing Lower Colorado River Water Shortages

by Jay Lund and Josué Medellin-Azuara

The lower Colorado River has been out of balance for about 40 years, using more water than has been available.  As their reservoirs empty, the three lower basin states, federal government, and water users are getting around to addressing this problem.  

The Colorado River reservoir system has immense storage capacity, about 4 times the river’s average annual historical inflow.  These Colorado River reservoirs (Lake Mead behind Hoover Dam and Lake Powell behind Glen Canyon Dam) have filled only once since their completion in 1963, in 1983.  Lower basin water demands have grown and a warming climate has reduced river flows and increased evaporation in the last century (Figures 1 and 2).  The reservoirs seem unlikely to refill anytime soon with this growing imbalance.  This long-term depletion of the basin’s surface reservoirs is paralleled by widespread overpumping of the region’s groundwater (Castle et al. 2014). 

Such immense reservoirs being full in 1983 might have prolonged inattention to fundamentals.  For managers, water users, and political leadership, starting with full reservoirs in 1983 has meant difficult policy choices could be postponed for almost two generations.

But with depletion of remaining reservoir storage progressing at a rate of about 3 million acre-ft per year, people are becoming concerned, and major decisions will need to be made within this or the next election cycle.  After 40 years, lower Colorado River basin reservoirs are falling to elevations where outlet structures can no longer release water or generate hydropower reliably.

Some imaginative and very expensive measures are being proposed (such as desalination, barring growth in Arizona, building new reservoirs, dragging icebergs upriver from Mexico, etc.).  However, the great majority of Colorado basin water use is for irrigated agriculture, mostly for animal feed, which seems far cheaper to fallow than the cost of most other proposed actions.

Figure 1. Lower Colorado River Reservoir Storage for 40 years (data from John Fleck)
Figure 2. Declining annual and 10-year average natural flow at Lees Ferry, Colorado River (maf/yr) (USBR data)
Figure 3: The Colorado River supplies seven states and Mexico

The Colorado River is a medium-sized river and the largest surface water supply for a region of more than 40 million people, 5.5 million acres of irrigated cropland, 22 federally-recognized tribes, and many national and state parks and wildlife refuges directly involving 7 western states and Mexico (Figure 3).

An agreement in 1922 and subsequent agreements and treaties allocate Colorado River water across 7 states and Mexico (Table 1) and set the basis for building the large lower basin reservoirs.  These initial allocations assumed unusually high average inflows from before 1922. At that time, the basin had more water than was being used by cities and agriculture, and seemed to allow for continued growth in water use.  The negotiations were eased somewhat by assuming a high average river flow and omitting tribal water rights and future reservoir evaporation.  Despite these flaws, this agreement helped hold the peace on the river for almost 100 years, aside from a few colorful episodes. 

A negotiated settlement for current shortages will be more difficult with today’s water scarce conditions, sizable water over-use in the lower basin, potential further water use growth in the northern basin, basin aridification from climate warming, and more pressing tribal water rights. 

Location1922 allocation (maf/yr)Recent actual use (maf/yr)
Average total water available16.412.5-14?
Upper basin7.54.8 (2018) **
   Colorado   3.86   2.4
   Utah   1.71   1.0
   Wyoming   1.04   0.4
   New Mexico   0.84   0.5
   Arizona   0.05   0.03
   Upper reservoir evap/losses   0.5
Lower basin states7.57.5 (2020)
   Nevada   0.3   0.3
   Arizona   2.8   2.8
   California   4.4   4.4
Reservoir evaporation1.5 (approximately)
Total Use~615.3
Annual Shortage-9 (surplus)3 (approximately)
Table 1: Colorado River water availability, allocations, and uses 1922 and recently * Under Minute 319 to the US-Mexico treaty, Mexico’s allocation can drop as much as 125 taf/yr in extreme droughts.
**USBR (2019) Provisional Upper Colorado River Basin 2016-2020.

Shortage is large

The Federal government has recently sought to reduce Colorado River water use by 2-4 maf/year.  A recent proposal from the states envisions a 2 maf/yr reduction to address depletion of water reserves.  The rough accounting in Table 1 indicates a sizable reduction of about 20%-30% of current water use is needed to balance supply and demand.

For the reservoirs to accumulate storage for future dry years, still greater cuts are needed.  The reservoirs are currently quite low, so it might be prudent to reduce water use by more so remaining supplies become more reliable.

Who’s most at risk?

The Colorado River supplies with water, energy, and recreation for tens of millions of people, trillions of dollars of economic activity, and some iconic landscapes and ecosystems.  So it is easy to panic about the Colorado River “running out of water.” 

Water use in the lower Colorado River basin is 70-80% for irrigated agriculture, with about four-fifths of this supporting feed crops for livestock.  Agricultural water use’s share of the economy is about 4%, and some fallowing historically has been a buffer for dry times. Urban water use is about 20-30% of Colorado River basin water use, but supports over 90% of the economy and people. Hydropower and recreation from the reservoirs mostly use water on its way downstream to consumptive agricultural and urban water users,

Few of these activities are at risk if the system is managed wisely, but hundreds of millions of dollars of losses are unavoidable, given the basin’s permanent water scarcity.  The costs of these shortages will be far greater ($ billions) and more widespread if the river’s available water is not managed and allocated well.

What to do?

Many options have been considered:

Clearly, expanding surface water storage capacity in the Colorado River basin will not improve its water supplies.  The lower basin lacks water to fill existing reservoir storage capacities even partially.  More storage capacity will not produce more water.

Desalinating seawater on the California or Mexican coasts has been proposed.  While technologically possible, coastal desalination’s roughly $2,000-$3,000/af cost greatly exceeds the economic value of most water uses.  The USBR’s less expensive brackish water desalination plant in Arizona has been mothballed for decades. 

Wastewater reuse already occurs throughout the basin, and almost no wastewater returns to the Sea of Cortez in Mexico.  So additional local wastewater reuse would merely take water from a current downstream user to expand a consumptive reuse of wastewater upstream.

Urban water conservation is important and will help some, but even draconian urban conservation cannot eliminate the need to permanently fallow irrigated croplands.  More importantly, urban conservation helps decouple water use from economic prosperity and population.   Southern California has imported about the same amount of water for decades, with a growing economy and population.  Conserve water in urban areas, but understand that this will not spare water-intensive agriculture from major reductions in water use. 

Nearly all the Colorado River Basin’s long-term water shortage will require permanent reductions in irrigated acreage.  Permanent reductions of 1-2 million acres of irrigated land will be needed in the lower basin states, about 20-30% of irrigated agriculture in the basin. 

An idea to facilitate permanent reductions in agricultural water use is a fee on all water use from the lower Colorado River to fund voluntary permanent buy-backs of agricultural water across the basin and some local costs of re-adjustment.  Well-established buyback schemes can help reduce cost of fixed reductions to all users.

California’s large agricultural water use has become a reserve water supply for southern California cities since the 1980s.  Since the 1980s, southern California cities have purchased water from more senior agricultural water-right holders, supplementing remaining agricultural profits with payments for fallowing and effective water conservation, often including some funds for local economic and social impacts.  The expansion of higher-valued crops in California and Arizona might increase costs of fallowing, but enough less profitable irrigated agriculture remains to achieve the needed water use reductions. 

Game theory

For most parties, the political game now is how to extract the most money from the Federal government and the most water from California so other lower-priority parties can reduce water use less.

Some interesting ideas for updating Colorado River allocations and their management include:

  • Firm up federal water rights for tribes outside of the 1922 agreement in a way that allows tribes to sell water anywhere so a) tribes win and their water claims are formally settled, b) a sizable water volume is shifted out of the 1922 agreement into a more flexible situation.  (This would disproportionately reduce California use and provide fungible long-term support for tribes.)
  • Do the same with evaporative losses from the reservoirs.  This was proposed recently by a group of six states (excluding California).
  • Human health and safety priority for some water, perhaps established at 55 gpcd (the current California drought standard) for residents depending on Colorado River water.  This would be a backstop for urban users, especially in Nevada and Arizona, supporting about 38% of current Phoenix and 50% of current Las Vegas water use.  The remainder of use could be from purchases of remaining allocations (perhaps across state lines) or other sources.
  • As a condition for Federal adjustment funds, states must allow their Colorado River water users to sell water to users in other states.  This allows the most important water uses in any state to be supplied reasonably, with compensation to users who give up more water than the law requires.

An important economic problem of how to minimize the economic and human impacts of unavoidable fallowing of irrigated lands.  Markets within states and perhaps across states can allocate remaining water to crops and people which provide greater economic benefits (and usually employ more and better paid workers). 

A subsequent environmental problem and opportunity for landowners, environmentalists, and local, state, and federal governments will be what to do with 1-2 million acres of permanently fallowed cropland. 

This need to reconfigure water management, agriculture, and land use in the lower Colorado River basin is not unique.  Many parts of the western US have overdeveloped water use and agriculture, worsened by climate change with deep environmental degradation.  Strikingly similar problems and solutions exist for addressing groundwater overdraft in California’s San Joaquin Valley, with very similar acreages of permanent fallowing needed.  Maintaining the Great Salt Lake in Utah, groundwater in Arizona, waterfowl and fish in the Klamath basin, and other terminal lake and playa and groundwater systems in the West (and globally, Lake Urmia in Iran and Aral Sea) all suffer from over-use of water mostly for irrigated agriculture. 

Human and ecological prosperity in the West with a changing climate and newer economic structures will require deep changes in water use and management similar to changes when these regions shifted from mining economies to agricultural economies.  Federal and state leadership, structure, and incentives can make these transitions occur more efficiently, quickly, and justly than will occur otherwise.

Jay Lund is a Professor of Civil and Environmental Engineering at the University of California – Davis.  Josué Medellin-Azuara is an Associate Professor of Civil and Environmental Engineering at the University of California – Merced. 

Further reading

USBR, Colorado River Basin Water Supply and Demand Study Executive Summary, US Bureau of Reclamation, December 2012

Castle, S., B. Thomas, J. Reager, M. Rodell, S. Swenson, and J. Famiglietti. “Groundwater Depletion during Drought Threatens Future Water Security of the Colorado River Basin.” Geophysical Research Letters 41, no. 16 (2014): 5904–11.

Bruce Babbitt (2022), “Department of Interior needs to review agricultural use of water amid negotiations for Colorado River cuts,” The Nevada Independent.

Porse, E. and S. Pincetl (2023), “Drought and the Colorado River: Localizing Water in Los Angeles,”, 8 January 2023.

Some other insightful media work:

Lower Colorado River Reservoir Evaporation the Focus of New Analysis.” 2022. KUNC. October 26, 2022.

Recent proposals from states:


Other states

John Fleck – Deadpool Diaries: The numbers in the states’ two proposals (1 February 2023)  

About jaylund

Professor of Civil and Environmental Engineering Director, Center for Watershed Sciences University of California - Davis
This entry was posted in Uncategorized. Bookmark the permalink.

20 Responses to Resistance is Futile – Agriculture is Key to Fixing Lower Colorado River Water Shortages

  1. bobnsf says:

    We have to eliminate all farming in the desert. California, Arizona, all of it. It’s insane. Farming needs to migrate to where there is water.

    • jaylund says:

      Probably not all farming, but probably the least valuable farming which consumes too much water. It will be about 20% less irrigated land in many arid areas.

    • Christo Gilberti says:

      Marc Reisner in Cadillac Desert noted ironically that while naturally irrigated fertile lands in the Tennessee Valley and environs were being flooded in the 30’s for the TVA, the largest water systems in the world were being constructed in the Central Valley of CA and the Southwest to convert semi- and desert lands to agriculture.

  2. Mathew McQuoid says:

    If you look into the amounts farming uses the same water today as in 1922how much have the cities grown let’s start with looking there

  3. E Nigenda says:

    Is there a way to keep hedge funds from buying land, mining the water and selling it to the highest bidder?

  4. Paul Zahn says:

    The real problem is never addressed: Overpopulation overusing limited resources.

    • Jeff says:

      Over population is literally not the problem. What Jay Lund points out has been stated in many articles regarding water use in the west, which is that about 80% is used for agriculture. That’s where most of the savings has to come from. Cutting the water used by people by 50% still would not solve our current water dilemma.

      • RobertL39 says:

        BUT the agriculture he’s talking about is used to feed people, either directly or after conversion to edible animal. Fewer people equals less need for growing feed stocks and other agriculture in difficult areas.

  5. Anna Warwick Sears says:

    I would love to read a blog post that went more into detail about this argument:

    “More importantly, urban conservation helps decouple water use from economic prosperity and population. Southern California has imported about the same amount of water for decades, with a growing economy and population.”

    I understand it, and agree with it, but have found it a bit of a hard sell sometimes in my region (western Canada).

  6. Jeff says:

    IMHO, this is maybe the most balanced article I’ve read on this topic. I will be linking to this article in the future. This is a complex topic that seems to be difficult for mainstream media to understand.

  7. Pingback: Western states play game of chicken over Colorado River - 24Talker

  8. Pingback: Skilling: Experts say Colorado river reservoirs unlikely to ever be completely refilled – WindyCityEvents

  9. Pingback: Les États jouent au poulet sur le fleuve Colorado - Nouvelles Du Monde

  10. Pingback: Western States Play Game of Chicken Over Colorado River – Pasadena Now

  11. Pingback: Western states play game of chicken over Colorado River » Holtville Tribune

  12. Pingback: Opinion: Western States Are Taking part in a Recreation of Hen With a Declining Colorado River – basternae

  13. Pingback: Western states play game of chicken over Colorado River - Calexico Chronicle

  14. James says:

    Really truthful, good read.
    With recent atmospheric rivers there must be a way to replenish ground water.
    Even taking water from Colorado River and storing underground in abundant years makes sense.

  15. Granite says:

    With a long view – dead pool (not the movie) will happen as early as July-August of 2023, although a colleague says it will likely happen 1-3 years later due to effective reservoir management to simply avoid dead pool. Bottom line – dead pool shall occur unless substantial reductions occur. Lets tighten the belt and not was our cars 3x per week in SoCal until more WWTP reuse and desal comes online sooner than later.

Leave a Reply