Which of the Five BOR Alternatives Survives Stress Testing?
The Bureau of Reclamation's Post-2026 Operating Guidelines Environmental Impact Statement evaluates five alternatives for governing Lake Powell and Lake Mead operations after the 2007 Interim Guidelines expire. Each alternative represents a different framework for allocating shortage, coordinating storage, and responding to drought triggers. The five alternatives span a range from operational continuity (No Action) to aggressive demand management and supply augmentation.
This investigation stress-tests each alternative by running 500 Monte Carlo traces across a full range of inflow scenarios from 8 MAF (severe drought, below 2020-2022 levels) to 16 MAF (wet conditions approaching 20th-century mean). The question is not which alternative performs best in average conditions — all five are designed for average conditions. The question is which alternative breaks first as inflows decline, and at what inflow level each alternative's breach probability crosses critical thresholds.
The five EIS alternatives: (1) No Action — extend 2007 Interim Guidelines; (2) Enhanced Coordination — improved Upper-Lower Basin cooperation; (3) Lower Basin Focus — accelerated Lower Basin demand management; (4) Maximum Flexibility — aggressive shortage sharing + demand management; (5) Supply-Driven — augment supply through new infrastructure (recycled water, desalination, transfers).
P(Breach) vs. Mean Inflow: When Each Alternative Breaks
The chart below shows the probability of a system-level storage breach (either Powell or Mead dropping below minimum power pool in a 30-year simulation) as a function of mean annual inflow. The vertical reference lines mark the current approximate mean (9 MAF, post-2000) and a severe drought scenario (8 MAF, below the 2020–2022 trough).
The separation between alternatives is stark. No Action and Enhanced Coordination have nearly identical response curves — both break near 9 MAF mean inflow. Lower Basin Focus performs modestly better. Maximum Flexibility and Supply-Driven maintain near-zero breach probability until inflows drop below 9 MAF, and even at 8 MAF, their breach probability remains well below 20%.
500 Monte Carlo traces per inflow scenario. 30-year simulation horizon. Breach = either Powell or Mead below minimum power pool in any year. Inflows drawn from synthetic ensemble; 9 MAF represents post-2000 observed mean; 8 MAF represents severe drought below 2020–2022 conditions.
The Compact Was Written for a Different River
The 1922 Colorado River Compact's Lee Ferry gauging station obligation requires the Upper Basin to deliver a 10-year running average of 75 MAF to Lee Ferry. At 7.5 MAF/yr Upper Basin use (the Compact's assumed UB consumption), this obligation is met when natural flows average 15 MAF/yr. The math worked in 1922.
At current Upper Basin consumptive use of approximately 4.5 MAF/yr, the delivery obligation requires approximately 10.5 MAF/yr of natural flow to be satisfied. The post-2000 mean is 12.4 MAF/yr — which appears sufficient. But 30-year Monte Carlo simulation reveals the problem: the distribution of 10-year running averages includes enough dry decades to push 30-year breach probability to 92.7%.
Compact breach = 10-year running average delivery to Lee Ferry falls below 75 MAF in any 10-year window during the 30-year simulation. Upper Basin development levels: Stable = 4.5 MAF/yr current UB consumptive use; Moderate Growth = 5.5 MAF/yr; Full Development = 6.5 MAF/yr (Upper Basin's full Compact entitlement). All runs at 12.4 MAF/yr mean inflow.
In the 2000–2009 decade — at current Upper Basin consumption (4.5 MAF/yr) — Lee Ferry delivery would have been 74.3 MAF, below the 75 MAF floor. Natural flow that decade was 119.3 MAF total, or 11.93 MAF/yr on average. The Compact obligation was already a near-miss in a recent historical decade.
The Lee Ferry Math: Then and Now
The 1922 Compact assumed Upper Basin states would develop their full 7.5 MAF/yr entitlement. At that development level, satisfying the 75 MAF/10-year Lee Ferry delivery requires average natural flows of at least 15 MAF/yr. The drafters used early 20th-century flow records showing means near 16–17 MAF/yr. This appeared conservative.
The 2000s were the decade the crisis became visible — and the delivery math was already failing at current demand levels. The 2007 Interim Guidelines and every subsequent negotiation were built around this existing condition without formally renegotiating the 75 MAF obligation itself.
The Policy Choice Is Mostly Irrelevant to Compact Risk
Model scope: this analysis uses annual delivery timesteps. The 92.7% figure captures 30-year cumulative breach probability but cannot resolve seasonal delivery dynamics within a year. The actual Compact breach probability could be higher if within-year delivery timing constraints matter — and they increasingly do as earlier snowmelt shifts the runoff window away from peak demand periods.
The EIS alternatives differ meaningfully on shortage distributions, operational flexibility, and response speed. Maximum Flexibility and Supply-Driven reduce system storage breach risk (Powell/Mead below power pool) substantially relative to No Action. But Lee Ferry compact breach is driven by the Upper Basin's consumptive use relative to natural flows — a structural condition that no operational guideline can directly address.
The No Action alternative's 2054 tipping point — the year at which median-scenario cumulative delivery to Lee Ferry first falls short of the 10-year running average floor — is not a forecast. It is a model output that illustrates the structural problem: under current demand trajectories and mean inflows, the Compact math does not close, regardless of operational sophistication.
Closing the Compact gap requires either reducing Upper Basin demand below current levels, augmenting supply above current mean inflows, or renegotiating the 75 MAF delivery obligation itself. None of those options appear in the EIS as modeled. The EIS optimizes within a legal framework that the analysis shows is already violated.
What This Analysis Does Not Resolve
30-year vs. 10-year obligation windows: The Lee Ferry obligation is a 10-year running average, not a 30-year cumulative. The 30-year simulation captures multiple obligation windows but treats them independently. A more sophisticated model would track whether a compact violation in one window resets or compounds the obligation.
Supply augmentation details: The Supply-Driven alternative's performance depends heavily on the volume and timing of new supply (recycled water, desalination, transfers). This analysis uses a generic 0.5 MAF/yr augmentation assumption. The actual EIS modeling uses project-specific timelines and costs that are not fully reflected here.
Climate trend: All Monte Carlo traces are drawn from the historical 1922–2022 distribution. Paleoclimate and climate model projections suggest declining mean inflows through mid-century. If the underlying distribution shifts downward, all breach probabilities increase and the distinctions between alternatives may compress further.
Institutional response: The model treats the alternatives as fixed operational rules. In practice, drought triggers, emergency declarations, and ad hoc shortage agreements (such as the 2023 IRA conservation agreements) can alter outcomes within any alternative. This analysis models stated policy, not actual negotiated outcomes.