Signals · Real Estate. A follow-up to “The Two Clocks of Colorado Springs Water” (July 6, 2026). All elevations are from USBR Glen Canyon hydrodata (site 919), cross-checked against Lighthouse's own append-only signal record. The two sources match to the foot.

In July we published a thesis about Colorado Springs water: the risk is seniority, not drought. We also stated, in print, exactly what would prove us wrong. Two conditions, either one sufficient:

“If Lake Powell ends summer 2026 above ~3,550 feet, or if the post-2026 rules adopted this year shield junior municipal transmountain diverters from compact-driven curtailment, the seniority thesis is overpriced and this piece overstated the risk.”

Summer is over. Here is the scorecard.

Arm one: the reservoir

Lake Powell ended September at 3,518 feet, thirty-two feet below the falsifier line.

The direction tells you more than the level. Powell fell more than ten feet between June and mid-September, through what is supposed to be fill season, the months when snowmelt runs the lake up. The July piece flagged a 2.5-foot fill-season decline in June as an early tell. The full season confirmed it: the system no longer fills in filling season. The lake ended summer about 28 feet above minimum power pool (3,490 feet), the level below which Glen Canyon Dam stops generating electricity.

Arm two: the rules

On August 21, the Interior Department issued the Record of Decision for post-2026 Colorado River operations. Three facts matter for the 719:

1. There was no basin consensus. The seven states could not agree, and the federal framework was imposed to keep the system operating, in two-year increments through 2036.
2. The cuts are real. Lower Basin apportionments drop 1.25 million acre-feet for 2027–28 (Arizona 760,000 acre-feet, California 440,000, Nevada 50,000).
3. Nobody upstream was shielded. The framework governs reservoir operations and Lower Basin deliveries. It does not, anywhere, protect junior municipal transmountain diverters from compact-driven curtailment. The question of who in Colorado goes short first if the compact ever binds remains exactly where it was: governed by the priority dates on the paper, with compact litigation now openly discussed in water-law circles.

The falsifier asked whether this year's rules would settle the seniority question. They left it open, and a question left open by negotiation tends to be answered in a courtroom, where the paper dates are the whole argument.

The honest accounting

Both arms of the falsifier failed to fire. The thesis stands. We want to be precise about what that does and does not mean. One season of being right is not proof. It is one observation that failed to kill the idea, and that is how this process is supposed to work. The claim was never “Powell will fall.” The claim was that Colorado Springs' real exposure lives in the seniority structure of its transmountain rights, on a clock measured in decrees and decades, and that day-to-day drought conditions are the wrong place to watch. Everything that happened this summer, a reservoir that could not fill in fill season and a rulemaking that ended without consensus and without shields, moved on that slow clock, in that direction.

The new falsifier, so the next review has a standard

Same discipline, next period:

If Lake Powell ends summer 2027 above 3,545 feet, or if the 2029–2030 operating guidelines (the framework's next increment) adopt protections that exempt post-compact municipal transmountain diversions from curtailment, the seniority thesis is overpriced and this follow-up overstated the risk.

The elevation feed updates on our cascade network every six hours. It is the same feed this verdict was scored against, and the numbers are there for anyone to check.

Sources: USBR Upper Colorado hydrodata, Glen Canyon Dam site 919 (daily elevations); USBR Post-2026 Record of Decision (Aug 21, 2026) and 2027–2028 Operating Guidelines; Lighthouse append-only signal record for infra_water_supply (June–October 2026).