Understanding the Permit. Understanding the Data.
A plain-language guide to the Project Jupiter air quality permit — what NMED is actually deciding, what the numbers mean, and what WhutNext can (and cannot) tell you.
What is this really about?
This is about an air quality permit. That is a yes-or-no question from the state: Can this company build its power equipment in a way that follows New Mexico’s air pollution rules?
The power plan uses many fuel cells (they turn natural gas into electricity) to run a large data center near Santa Teresa. NMED is not voting on whether you like data centers, AI, or jobs. It is checking the air pollution side of this one power project. Dive deeper on fuel-cell scale and what is (and isn’t) proven.
Two different “cleaner” stories (read this first)
You might hear “92% cleaner” and “10 million tons of pollution” in the same conversation. Those are both real talking points — but they measure different things. This table is Oracle/YGI’s comparison of the new fuel-cell plan to the old gas-turbine plan they withdrew in April 2026. NMED has not confirmed these figures.
| Pollution type | In plain English | Applicant says: change vs old plan |
|---|---|---|
| NOx | Smog-forming gas — big deal near Sunland Park’s ozone problems | ↓ ~92% |
| Soot (PM) | Tiny particles in the air | ↓ ~83% |
| Carbon monoxide | Poisonous gas from incomplete burning | ↓ ~67% |
| VOCs | Chemical vapors that can form smog | ↓ ~38% |
| CO₂ / greenhouse gas | Climate pollution — still huge in absolute terms | ↓ ~21% but still ~10.1 million tons per year |
~29% less on this scale — but 10.1 million tons is still an enormous amount.
Sources: Oracle’s July 2, 2026 NMED letter (percent reductions vs January 2026 turbine filing); ~14M / ~10.1M tons CO₂e from press reporting on the updated permit. WhutNext has not independently verified these numbers.
A WhutNext question — why not this plan from the start?
If the fuel-cell design is so much better on smog and water, why lead with gas turbines? That is a fair trust question — not a partisan attack. Here is what the public record shows.
What happened (timeline):
- Fall 2025: Developers filed air permits for on-site gas turbines — a common data-center approach (fast, proven at huge scale, behind-the-meter).
- Dec 2025: NMED said those applications were incomplete, partly over nitrogen oxides (NOx).
- Winter–spring 2026: Thousands of public comments, lawmakers asking for a hearing, pipeline permits denied, lawsuits filed.
- April 2026: Developers withdrew the turbine permits and filed the fuel-cell plan instead.
What the builders say
- They are “listening” and improving the project
- Fuel cells were not viable at this scale until the Bloom partnership matured
- The new design is a meaningful upgrade for local air and water
What skeptics hear
- Start with the easiest path; upgrade only when permits and politics stall
- “Cleaner” mostly fixes local air metrics, not the core fossil-fuel choice
- Switching designs also reset the permit clock after heavy opposition
What they are still not doing (as of this filing):
- Not grid-powered — still building a private ~2.45 GW gas-fueled
, not plugging into utility renewables
“microgrid”
Marketing calls it a microgrid, but the permit is for ~2.45 GW. Compare that to New Mexico’s grid:
- ~11.9 GW NM net summer plant capacity (EIA 2024) — Jupiter is about one-fifth of that
- ~3.5 GW average statewide electricity load — Jupiter is roughly 70% of that
- ~2.3 GW El Paso Electric summer peak, southern NM (2024) — about the same size as this plant alone
- Not off fossil gas — fuel cells still use natural gas today (~10.1M tons CO₂e/year in filings)
- Not small — still one of the largest behind-the-meter power projects proposed in the U.S.
- Not water-free — large one-time fill plus ongoing maintenance water; cooling is a separate water story
- Not waiting for full public review — campus construction has moved ahead while air permits were contested and revised
What NMED can look at here
- How much pollution the fuel cells would put in the air
- Thousands of stacks and whether controls are enough
- Whether the math and models in the permit make sense
- Permit rules, monitoring, and operating limits
What this permit is not
- A popularity vote — more letters does not automatically win
- A jobs contest — “4,000 construction jobs” is not an air rule
- A full veto on the whole data center campus
- The main place to decide water rights (water matters, but it is a different lane)
Has NMED decided if the project meets air quality standards?
No — not yet. The permit is still under review. The public comment period is closed, but NMED has not issued a final yes or no.
Press reports suggest NMED may decide around late summer 2026. Until then, the project is not approved on air-quality grounds — no matter how many comments were filed.
What would a “yes” require?
NMED staff and engineers review the application — not the comment count. For approval, the paperwork would need to show things like:
- Pollution limits are met — emissions of smog-forming gases (like NOx), soot, and other regulated pollutants stay within the rules
- Air quality math checks out — models show the project would not push local air past health-based standards
- Controls and monitoring are adequate — thousands of fuel-cell stacks would need enough pollution controls, with ways to enforce limits
- Extra care near Sunland Park — part of Doña Ana County already struggles with ozone (smog), so new NOx sources face tougher review
What this page cannot tell you: WhutNext analyzed public comments, not NMED’s engineering file. The applicant says the fuel-cell plan is much cleaner than the old gas-turbine plan. Critics say it would still release very large amounts of greenhouse gas and raise stack-and-control concerns. Only NMED can decide if the application meets New Mexico’s air rules.
How the comment numbers connect to the permit
Most of the huge total is the same letter over and over with different names on line one. The full report breaks down templates, charts, and timeline.
Of the “other” comments — what connects to the permit?
We tagged 5,725 non-template comments. Here is how they group (automated keyword scan — directional, not official):
- In scope 1,318 (23.0%) — air permit issues NMED can weigh
- Adjacent 2,108 (36.8%) — water, health, hearings (related, but not the core air test)
- Out of scope 331 (5.8%) — jobs, generic support/oppose
- Unclear 1,968 (34.4%) — too short or hard to classify
Top air-permit themes in non-template comments
- 1,122 Source configuration (fuel cells / stacks)
- 572 Emissions quantities & pollutants
- 158 Air quality modeling & standards
- 62 Major source / NSR framing
Takeaway: The docket is enormous, but the part that speaks directly to this air permit is a smaller slice — mostly questions about fuel cells, stacks, emissions, and controls. NMED will still read everything, but it answers the permit on rules and science, not on headcount.