Data-Driven Climate Analysis
The American West's
Water Is Already Gone
A 100-year predictive model grounded in peer-reviewed climate science reveals the Colorado River crisis isn't approaching — it's here. The structural deficit exists today. The question isn't whether to act, but whether we've already waited too long.
Note on scenarios: SSP5-8.5 represents a high-emissions upper bound. Most researchers now consider SSP2-4.5 to SSP3-7.0 more likely given current policies. We show both to bracket the range. Even the moderate scenario demands urgent action.
4–6
MAF/year structural deficit on the Colorado River
99%
Probability of major drought within 100 years
1,200
Years since a drought this severe (2000–2021)
42%
Of current drought severity caused by human warming
01 — The Structural Deficit
Colorado River Flow Projections
The legal allocation of 18.3 MAF/year has never matched reality. The river's 500-year average is just 13.2 MAF. Climate change is widening a gap that was structural from the start.
Projected Flow vs. Legal Allocation
The deficit is structural and permanent. The Colorado River Compact was negotiated in 1922 during the wettest period in 1,200 years. Even the 500-year tree-ring average (13.2 MAF) falls 5.1 MAF short of legal commitments. Climate change is accelerating the problem, not creating it.
02 — The Widening Gap
Supply-Demand Deficit
The gap between legal promises and physical reality widens every decade as warming reduces snowpack and increases evapotranspiration. Flow drops ~6.5% per degree Celsius of warming.
Annual Water Deficit (MAF/Year)
03 — The Probability
Cumulative Drought Risk
Derived from Ault et al. 2016 and Cook et al. 2015 megadrought risk frameworks applied to CMIP6 temperature trajectories.
Cumulative Probability of Major Drought
Probability by Timeframe & Severity
| Window |
Moderate Emissions (SSP2-4.5) |
High Emissions (SSP5-8.5) |
| Moderate |
Major |
Extreme |
Moderate |
Major |
Extreme |
| Next 10 years |
48.2% | 29.0% | 12.8% |
53.1% | 33.5% | 18.3% |
| Next 25 years |
78.5% | 59.4% | 30.2% |
84.3% | 68.2% | 43.2% |
| Next 50 years |
95.2% | 86.5% | 55.6% |
98.1% | 94.3% | 75.6% |
| Next 75 years |
99.0% | 96.0% | 72.4% |
99.8% | 99.3% | 91.3% |
| Next 100 years |
99.7% | 98.9% | 82.6% |
99.99% | 99.9% | 97.1% |
Moderate drought: 5+ consecutive below-normal years. Major drought: 10+ year megadrought. Extreme drought: unprecedented severity exceeding 2000–2021 levels. Each row reads left→right from most likely to least likely within each emissions scenario.
04 — The Backup Is Failing
Aquifer Depletion
Groundwater is the backup when surface water fails. But the Ogallala — the largest aquifer in the US — is being depleted at ~11 km³/year. The southern sections face functional exhaustion within decades.
Ogallala Aquifer Remaining Storage
The southern Ogallala is a non-renewable resource. In the Texas panhandle and western Kansas, recharge rates are negligible compared to withdrawal. When it's gone, it's gone for geological timescales. Parts of the southern High Plains could lose irrigated agriculture entirely by 2050–2070.
05 — Current Efforts vs. The Need
Is What We're Doing Enough?
Active programs are producing real results — California recharged 7.4M AF over 2022–2024. But most recharge depends on wet years. During drought, when you need it most, the spigot runs dry.
Current Efforts vs. Projected Deficit
The drought paradox: Most recharge capacity depends on surplus water that disappears during the droughts it's meant to buffer against. California's 2.5M AF/year of recharge vanishes in dry years. The AWBA can't bank water it doesn't receive. Arizona's 4.47M AF strategic reserve covers just 9 months at the current deficit rate.
What the current effort gets right: Las Vegas cut per-capita use 40%, recycles 99% of indoor water, and banned non-functional turf. Orange County proved potable reuse at 130 MGD. Idaho's ESPA has recharged 2.3M AF since 2014. Salt Lake County is coming online. These are proofs of concept — they need to be replicated everywhere, at 5–10x scale, within a decade.
06 — Interactive Model
How Much Investment Does It Take?
Drag the slider to explore how different levels of federal and state investment change water stress outcomes across the West by 2100. Watch the map transform in real time as capital flows to the most cost-effective technologies first.
Total New Infrastructure Investment
$0$25B$50B$75B$100B
0
States below moderate stress
New Capacity Delivered Across All States
Existing capacity: 0.85 MAF/yr
+ 0 MAF/yr new
0
~14 MAF/yr total needed across 17 states
Where the Money Goes
Aquifer Recharge
0
MAF/yr new capacity
Ag Conversion
0
MAF/yr water saved
Water Recycling
0
MAF/yr new capacity
Desalination
0
MAF/yr new capacity
Projected Water Stress by State (2100)
Hover any state for details. Scores: 0–20 low, 20–40 moderate, 40–60 high, 60–80 severe, 80–100 catastrophic
Investment Required by State to Reach Safe Levels
Each state needs its own physical water capacity. The total is the cumulative sum across all 17 states.
Adjust the slider above to see states turn safe in real time.
| State |
Baseline |
Current Score |
Capacity Needed (MAF/yr) |
National $ Threshold |
Primary Strategy |
| Cumulative total (all 17 states): |
|
|
|
Key insight: Drag the slider to ~$65B and watch every state on the map turn green — that's the cumulative investment to deliver ~14 MAF/yr of new capacity across all 17 western states. That's less than a single year of federal highway spending ($67B in 2024). California ($22B) and Texas ($18B) are the two largest line items, driven by expensive coastal desalination. The technologies are proven — the price tag is known — what's missing is the political decision.
07 — The Solutions Exist
Water Storage & Conservation Technology
Ranked by cost-effectiveness. The most impactful technologies are also the cheapest and most proven.
Cost per Acre-Foot by Technology
Deploy Now
Managed Aquifer Recharge
$150–500 / acre-foot
Store surplus surface water underground for later recovery. Arizona's Water Bank has banked 4.47M AF. Recovery rates 60–95%. The single most cost-effective large-scale storage technology.
Deploy Now
Water Recycling
$1,000–1,800 / acre-foot
Advanced treatment of wastewater to drinking standards. Orange County's GWRS produces 130 MGD. Singapore gets 40% of supply from recycling. Output exceeds drinking water quality.
Deploy Now
Smart Irrigation + Curtailment
Varies by crop
Converting flood irrigation (40–65% efficient) to drip (85–95%). Agriculture uses 70–80% of Western water. Must pair with curtailment mandates to prevent Jevons Paradox.
Build Within 10 Years
Seawater Desalination
$1,800–2,500 / acre-foot
Proven at national scale — Israel desalinates 80% of domestic water. Energy-intensive but improving. Coastal-only. Strategic backup, not primary source.
Build Within 10 Years
Stormwater Capture
$500–1,500 / acre-foot
Capture urban stormwater and route to aquifer recharge basins. LA's master plan targets 132,000 AF/year. Solves flood management and water supply simultaneously.
Policy Required
Colorado River Compact Reform
$0 (political capital)
Renegotiate allocations to match reality (~12–13 MAF). Regulate all groundwater. Create ag water markets. Tiered pricing. Most impactful lever, hardest to pull.
Build Within 10 Years
Brackish Desalination
$500–1,200 / acre-foot
Treating inland brackish groundwater. Cheaper than seawater desal. Available without ocean access. El Paso's KBH plant produces 15,000 AF/yr — largest inland desal globally.
Limited Viability
Atmospheric Water Generation
$3,000–10,000+ / acre-foot
Thermodynamically constrained: works worst where needed most. Produces liters/day, not the millions of acre-feet required. Not a serious bulk water solution.
08 — The Urgency
Why Planning Must Start Now
This is not a future problem. Every key indicator shows the crisis is underway.
-
The structural deficit exists today
The Colorado River is overallocated by 4–6 MAF/year right now. The post-2026 operational framework — being negotiated by the Bureau of Reclamation and basin states — is the most critical water policy event in a generation.
-
The megadrought is ongoing and unprecedented
2000–2021 was the worst 22-year drought in 1,200 years. 42% of its severity is directly attributable to human-caused warming, which will only accelerate.
-
Population growth collides with shrinking supply
Arizona, Nevada, Utah, and Colorado are projected to grow 30–60% by 2050 while their primary water source declines.
-
Groundwater — the backup — is being drained
The southern Ogallala faces functional exhaustion by 2050–2070. Arizona's aquifers are being pumped harder as Colorado River curtailments bite.
-
Infrastructure takes decades to build
Desalination plants, aquifer recharge systems, and recycling facilities take 10–20 years from planning to operation. Starting in 2040 means capacity arrives after the worst impacts hit.
-
Every year of delay increases cost exponentially
Emergency water costs 5–10x planned infrastructure. Proactive MAR costs $150–500/AF. Crisis-mode desal costs $2,500+/AF. The math is unambiguous.
09 — What You Can Do
A Plan of Action
The data is clear. The technology is proven. The cost is known. What's missing is organized demand for political action. Here's how to turn this analysis into outcomes.
For Individuals & Households
-
Contact your state legislators about water infrastructure funding
State legislatures control water policy. Find your representatives at openstates.org and demand they fund MAR, recycling, and conservation programs. The federal Infrastructure Investment and Jobs Act allocated $8.3B for western water — your state needs to apply for and deploy those funds.
-
Support the Colorado River post-2026 operational framework
The Bureau of Reclamation is finalizing new operating rules for the Colorado River. Public comment periods matter. Follow updates at usbr.gov/ColoradoRiverBasin and submit comments supporting allocations based on actual river flows (~12–13 MAF), not the fictional 1922 numbers.
-
Reduce your household water footprint
Replace ornamental turf with native landscaping. Install low-flow fixtures. Fix leaks. Residential use is ~10% of total western water, but per-capita reduction demonstrates public will and enables policy. Las Vegas proved that cutting per-capita use 40% is possible without reducing quality of life.
-
Vote on water bonds and infrastructure measures
Water infrastructure ballot measures regularly appear in western states. They fund the exact MAR, recycling, and stormwater systems this analysis shows are needed. Research water measures before every election.
For Policymakers & Water Managers
-
Fund Managed Aquifer Recharge at 5–10x current scale
Arizona's Water Bank, Idaho's ESPA, and California's Flood-MAR have proven the model. Every viable aquifer basin in the West needs a recharge program. This is the highest-ROI water investment at $150–500/AF — a fraction of the cost of emergency measures.
-
Mandate indirect potable reuse for cities over 100K
Orange County's GWRS proves it works at scale. Recycled water exceeds drinking water quality standards. Every major western city should be recycling 80%+ of wastewater by 2035. The technology exists — permitting and public education are the bottlenecks.
-
Reform agricultural water rights with market mechanisms
Agriculture uses 70–80% of western water. Mandatory conversion from flood to drip irrigation, paired with water markets that let farmers sell conserved water, can free up 2–3 MAF/yr without destroying farm economies. Pair with curtailment mandates to prevent Jevons Paradox.
-
Regulate all groundwater pumping — end the exemptions
California's SGMA is a start but covers only designated basins. Arizona exempts rural wells. Texas has no statewide regulation. Unregulated pumping is depleting aquifers that took millennia to fill. Universal metering, reporting, and allocation are non-negotiable.
-
Build strategic desalination capacity on both coasts
Not as a primary source, but as drought-proof backup. Israel desalinates 80% of domestic water. California and Texas should have 1–2 MAF/yr of desal capacity as insurance against the worst-case scenarios shown in this analysis.
For Researchers & Journalists
-
Scrutinize and improve this model
This is a simplified scenario model for public communication, not a substitute for CMIP6 ensemble analysis. The source code and data are open at github.com/kirkwarren/water-scarcity-west. File issues, suggest corrections, and help make the data more accurate.
-
Cover the Colorado River renegotiation as the existential story it is
The post-2026 operational framework determines water allocation for 40 million people. It deserves front-page coverage proportional to its impact — more consequential than most political stories that dominate the cycle.
-
Investigate state-level groundwater data gaps
Many western states lack comprehensive groundwater monitoring. The data gaps hide the true rate of aquifer depletion. GRACE satellite data shows the macro trend, but state-level reporting is inconsistent.
The bottom line: ~$65 billion in proven water infrastructure — less than one year of federal highway spending — gets every western state to safe water stress levels. The technologies work. The math works. The only variable is whether we build it before the crisis forces us to, at 5–10x the cost, in the middle of a drought. Share this analysis. Contact your representatives. Demand a plan.
Sources & Methodology
- Williams, A.P. et al. (2022) — "Rapid intensification of the emerging southwestern North American megadrought in 2020–2021." Nature Climate Change.
- Cook, B.I. et al. (2015) — "Unprecedented 21st century drought risk in the American Southwest and Central Plains." Science Advances.
- Ault, T.R. et al. (2016) — "Relative impacts of mitigation, temperature, and precipitation on 21st-century megadrought risk." Journal of Climate.
- Udall, B. & Overpeck, J. (2017) — "The twenty-first century Colorado River hot drought." Water Resources Research.
- CMIP6 Multi-Model Ensemble — SSP2-4.5 and SSP5-8.5 scenarios.
- Arizona Water Banking Authority — Annual Reports. 4.47M AF banked.
- Idaho Water Resource Board — ESPA Aquifer Stabilization. 2.3M AF since 2014.
- California DWR — Flood-MAR Program; 7.4M AF managed recharge 2022–2024.
- Metropolitan Water District of Salt Lake & Sandy — MAR Project.
- City of Aurora, CO — Prairie Waters Project.
- Governor of California (March 2026) — Groundwater recharge report.
Methodology: Simplified scenario model for public communication. Flow projections use Udall & Overpeck (2017) temperature-sensitivity applied to CMIP6 trajectories. Drought probabilities from Ault et al. (2016) and Cook et al. (2015). For operational decisions, consult Bureau of Reclamation's full CMIP6 ensemble.