the colorado river runs out of rules in 2026 — here's what actually fixes it

On December 31, 2026, the Colorado River doesn't run dry. Its rules do.

The three agreements that decide who gets cut, by how much, and when — the 2007 interim guidelines, the 2019 drought contingency plans, and Minute 323 with Mexico — all expire at midnight. Seven states have spent two years failing to agree on what replaces them. Miss the federal deadline, and the Bureau of Reclamation writes the rules for them.

If you run a water utility, sit on a county commission, hold water-dependent property, or invest anywhere in the Southwest, 2027 is a blank page. This is the plain-language decoder: what actually expires, what the government's options really do, and why every one of them still leaves the West fighting over a shrinking river — plus the one move that grows it back.

what actually expires on december 31, 2026

First, what doesn't expire. The 1922 Colorado River Compact — the document that split the river between an Upper Basin (Colorado, Wyoming, Utah, New Mexico) and a Lower Basin (Arizona, Nevada, California), 7.5 million acre-feet each — is permanent. So is the 1944 treaty guaranteeing Mexico 1.5 million acre-feet a year. Those stay.

What expires is the crisis-era operating manual layered on top: the rules written over the last two decades to manage shortage as the reservoirs fell.

What expires in 2026 is not the Colorado River or its foundational law — it's the temporary shortage-sharing rules that tell each state how much to give up, and when.

what happens on january 1, 2027

Your tap still works on January 1. This is not a shutoff. But the largest reservoir system in the country loses its shared operating logic at the worst possible moment, and two paths open:

1. The states agree. Reclamation finalizes a successor framework — a record of decision was targeted for late 2026 — and new rules take effect.
2. The states deadlock. The federal government imposes operating rules, and litigation becomes near-certain, including the long-feared "compact call."

Here's where the system actually sits as those rules expire.

Lake Powell's minimum power pool is 3,490 ft — below it, Glen Canyon Dam stops generating hydropower for roughly 5.8 million people. Dead pool, where water can no longer pass through the dam at all, sits near 3,370 ft at Powell and 895 ft at Mead. To hold Powell up, upstream reservoirs are already being drained: Flaming Gorge is releasing up to 1 million acre-feet through April 2027. Under the current tier-1 shortage, Arizona gives up 512,000 acre-feet (−18%), Nevada 21,000 (−7%), and Mexico 80,000 (−5%). California, holding the most senior rights, takes no mandatory tier-1 cut.

the DEIS: five ways to divide the same shrinking river

In January 2026, Reclamation released its post-2026 environmental impact statement (DEIS) — roughly 1,600 pages, five alternatives, and no named preferred option. The comment window closed in March 2026 with more than 18,000 comments filed.

The alternatives range from continuing today's rules (unsustainable) to spreading cuts proportionally across both basins, with tribal water rights complicating every version. Read past the page count, though, and every alternative shares one feature.

the trap nobody at the table can escape

The 1922 compact promised about 19 million acre-feet a year. The river now delivers 10 to 12, and the 24-year megadrought that scientists increasingly call aridification — a permanent drying, not a dry spell — is pushing that lower.

That gap is the whole story. Tiers, cuts, curtailments, and compact calls are all the same move performed with different weapons: dividing a smaller pie.

A compact call — the Lower Basin legally forcing the Upper Basin to curtail use to meet the delivery obligation at Lee Ferry — has never been tested in court. Even if someone wins it, the prize is a larger share of a shrinking flow. It moves water between users. It does not make more.

grow the source, don't just fight over the tap

Here is the move nobody at the negotiating table is empowered to make.

Roughly 90% of the Colorado's water starts as snow and rain in the forests, meadows, and alpine headwaters of the Upper Basin. That's the factory floor. When those source lands burn, when beetle-killed forests stop shading the snowpack, when meadows and wetlands that meter out snowmelt are lost, less water reaches the river — no matter how the lawyers slice the allocation downstream.

The reservoir level is a symptom. The headwaters are the supply. Protecting and restoring source lands is the only lever that adds water to the system instead of moving it between users. Healthier headwaters hold snowpack longer, cut evaporation and catastrophic wildfire, and improve how efficiently snowmelt becomes streamflow.

which acres actually matter — the spillover question

"Protect the watershed" is 246,000 square miles of advice. The useful question is narrower: which specific acres, feeding which specific users, are worth protecting first?

Not all source land pulls the same weight. A small, identifiable minority of parcels — certain headwaters positions, riparian corridors, the exact land that feeds a given intake — carries most of the downstream benefit. A public, pre-registered network analysis of the Upper Colorado corridor, built with BASIN, found the load-bearing parcels clustered tightly in the Eagle headwaters: a keystone fraction, not the whole map.

That work is honest about its limits — it's screening-grade, not investment-grade. It tells you where to look, not what you'll earn. But it answers the question allocation fights never ask: which upstream acre decides your downstream water. Think of it the way a business thinks about a supplier. You would never leave your single most important supplier uncontracted, unpaid, and unmaintained. For 40 million people and $1.4 trillion in annual economic activity, the headwaters are exactly that supplier.

who pays — and why it isn't charity

The cities, utilities, farms, and property owners who lose the most when the source fails are the natural payors for protecting it. That's the core idea behind ensurance: fund protection upfront, as an investment held against future loss — not as a donation, and not as insurance paid out after the damage is done.

The economics are not speculative. New York City protected its Catskills watershed for a fraction of the cost of the filtration plant it would otherwise have had to build — source protection came in multiples cheaper than concrete. The same avoided-cost logic applies to Colorado River headwaters: the cheapest acre-foot is often the one you never lose.

Two instruments carry it. A certificate (specific ensurance) funds a single named source-water asset directly. A coin (general ensurance) funds protection across the basin. Both route proceeds to the places and stewards doing the actual work — and both are held, not spent.

Ensurance lets the beneficiaries of a watershed — cities, utilities, farms, property owners — fund its protection upfront and hold it as an investment, instead of paying for the damage after the source fails.

what this means for you

The post-2026 rules will decide how a shrinking river gets divided. They will not decide how much river there is. That's the lever this series is about — and it's available regardless of which DEIS alternative wins.

• water utilities & districts — source-water protection is often cheaper than the next treatment plant or pipeline, and it hedges the supply your rate base depends on. explore how watershed value is measured →
• governments & regional collaboratives — protection funded upfront survives allocation politics and grant cycles alike. see the protocol →
• investors & capital providers — the durable premium in an over-allocated basin accrues to protected supply, not contested paper rights. see specific ensurance →
• landowners & farms — the land that makes water can be paid to keep making it. talk to someone who can help →

frequently asked questions

what happens to the colorado river after december 31, 2026?

The three crisis-era operating agreements — the 2007 interim guidelines, the 2019 drought contingency plans, and Minute 323 with Mexico — expire. If the seven basin states agree on a successor framework, new rules take effect; if they deadlock, the federal Bureau of Reclamation imposes operating rules, likely triggering litigation. Water keeps flowing, but the rules for sharing shortage reset.

will the federal government impose water cuts if the states can't agree?

Yes. Reclamation controls dam operations and has signaled it will set post-2026 rules itself if the states miss the deadline. A federally imposed framework would almost certainly be challenged in court, which is why a negotiated deal is preferred by nearly everyone at the table.

what is dead pool?

Dead pool is the reservoir elevation below which water can no longer pass downstream through a dam's outlets — roughly 895 ft at Lake Mead and 3,370 ft at Lake Powell. Above it but still critical is minimum power pool (about 3,490 ft at Powell), below which the dam stops generating hydropower.

what is a colorado river compact call?

A compact call is a legal demand by the Lower Basin that the Upper Basin curtail its use to meet the 1922 compact's delivery obligation at Lee Ferry (framed around 82.5 million acre-feet over any 10 years). It has never been tested. Even if invoked and won, it reallocates water between users — it does not create more.

does protecting the watershed actually add water?

It won't reverse aridification, but healthy headwaters forests, meadows, and wetlands hold snowpack longer, reduce evaporation and catastrophic wildfire, and improve how efficiently snowmelt becomes streamflow. Source protection is the only lever that adds to supply rather than reassigning it — which is why it outlasts every allocation fight.