Scale
Who depends on the Colorado River?
Approximately 40 million people, 5.5 million acres of farmland, 30 Tribal Nations, seven U.S. Basin states, and two Mexican states depend on the system.1
Reclamation also reports that Colorado River hydropower produces more than 8 billion kilowatt-hours annually—enough for approximately 700,000 homes—and that the river supports wildlife refuges, recreation areas, and national parks.1
The basic unit
What is an acre-foot?
One acre-foot is about 326,000 gallons: enough water to cover one acre of land one foot deep.2
Water managers use acre-feet because river flows, reservoir storage, and annual deliveries are too large to describe conveniently in gallons. “MAF” means one million acre-feet.
The full water budget
Who consumes Colorado River water?
For 2000–2019, a peer-reviewed basin-wide study estimated 52% of total consumption from irrigated agriculture, 18% from municipal, commercial, and industrial use, 11% from reservoir evaporation, and 19% from riparian and wetland vegetation.3
These are basin-wide averages over two decades, not every year’s exact balance. They include water consumed by people and by managed river landscapes. Different datasets may use different boundaries and accounting conventions, so their percentages should not be mixed without checking definitions.
Food and working lands
Does agriculture use most of the water?
Yes. Irrigated agriculture is the largest direct consumptive-use sector. The same 2000–2019 study estimated alfalfa alone at 27% of total Basin consumption and other hay at 4%.3
That fact is a reason to invest with agricultural producers—not to erase agricultural communities. Farms produce food, sustain rural economies, preserve open land, and often hold senior water rights. Because agriculture is the largest flexible use, voluntary agricultural participation can also produce meaningful system savings.
The practical conclusion: cities cannot stabilize the system alone, and agriculture should not be expected to provide water for free. Conservation has to offer enough value for producers and communities to choose it.
Municipal use
Could cities solve the problem by conserving?
Cities must conserve, but municipal, commercial, and industrial use represents a minority of total Basin consumption—about 18% in the 2000–2019 water budget.3
Urban conservation, recycling, leak reduction, efficient landscaping, and growth planning all matter. But even deep urban reductions cannot replace Basin-wide action across agriculture, reservoir operations, environmental needs, and interstate commitments.
System losses
How much water is lost to evaporation and vegetation?
The basin-wide study estimated reservoir evaporation at 11% of total consumption and riparian and wetland evapotranspiration at 19%.3
Not all of this is avoidable or undesirable. Reservoirs provide storage, hydropower, flood control, recreation, and operational flexibility. Riparian areas and wetlands support habitat and river health. Good accounting makes these uses visible; it does not assume all can be eliminated.
The Law of the River
How much water was allocated?
The 1922 Compact apportioned 7.5 MAF annually to each of the Upper and Lower Basins. The 1944 treaty provides 1.5 MAF annually to Mexico, subject to treaty terms.2
Those headline numbers are not a simple guarantee that 16.5 MAF will arrive every year. The system is governed by a layered “Law of the River,” hydrology varies, evaporation and other losses matter, and individual rights and obligations differ by state, priority, contract, decree, and treaty.
Balancing the system
How much conservation is needed?
In 2022, Reclamation called for 2–4 MAF per year of additional conservation to protect critical reservoir elevations.4 That was an urgent near-term planning range—not a timeless estimate of the river’s fixed annual deficit.
A durable program should set an annual acquisition target from observed inflow, forecast uncertainty, reservoir storage, legal delivery requirements, prior conservation, and an explicit safety margin. It should buy less in wet periods and more when risk rises.
Why a reserve matters: waiting until reservoirs approach critical elevations makes conservation more expensive, more disruptive, and harder to verify. Purchasing options and savings in advance creates room to manage dry years.
Accounting
What counts as water saved?
The relevant measure is a verified reduction in consumptive use—not simply a smaller diversion.
If a farm diverts less but downstream return flow falls by the same amount, the system may gain little or nothing. Creditable conservation needs a defensible historical baseline, measurement of actual conditions, adjustments for weather and crop changes, return-flow protection, conveyance-loss accounting, and independent review. This is the core design requirement of the proposed Conservation Bank.
Evidence
Primary sources
- U.S. Bureau of Reclamation, Colorado River Post-2026 update — people, agriculture, Tribal Nations, states, parks, and hydropower.
- U.S. Bureau of Reclamation, Colorado River water contracts and use — Compact apportionments, Mexican treaty delivery, and acre-foot conversion.
- Richter et al., Communications Earth & Environment (2024) — peer-reviewed complete water budget for 2000–2019.
- U.S. Bureau of Reclamation environmental assessment — documentation of the 2022 call for 2–4 MAF annually of near-term conservation.
- U.S. Bureau of Reclamation, Consumptive Uses and Losses Report — consumptive-use definition, return flows, evaporation, and accounting methods.
- U.S. Geological Survey, Water Use and Trends in the Colorado River Basin — sector definitions and historical water-use dataset.
- U.S. Bureau of Reclamation, Upper Basin Consumptive Uses and Losses reports — current federal accounting series.
Percentages are rounded. This page is public education, not a legal determination of any water right or user’s consumption.