The record
Every hour priced, graded, and published, with the holds shown and the reasons attached. The rare passing hour is the halo tier, not the headline.
The meter, live
The year against the gate, hour by hour.
Every cached hour of the meter's trailing year on MISO Central, graded through the same engine that mints. The hour-mass sits far above the 100 gCO₂/kWh ceiling; the gate did not open once. The everyday output is the graded record and the clean certification unavailable signal.
REAL REAL · the meter's own MISO market-feed day cache, about 24 hours delayed, never real-time.
Median hour 527.6 gCO₂/kWh; the cleanest hour of the year stayed 174 g above the ceiling. Peak restraint (demand above 75 GW whole-MISO) would have held 53.2% of hours on its own; a missing reading fails closed. On this window the clean gate alone set the count.
MODELED the facility water leg is MODELED today (published WUE); 1.8 L/kWh is the labeled interim default ceiling, in force where no basin prescription exists yet.
Where the same gate opens routinely, the board shows it. The difference is policy, not technology. The national board, below →
Every grid, one question
Dozens of balancing authorities run the lower 48, and the same federal feed reports every one of them. Aquadrio reads that feed on the hour and asks each grid the same question the meter asks its own: is the gate open right now? Most hours, almost everywhere, the answer is no. Where it is yes, the difference is policy, not technology.
Reading the board…
Reading the board
Each tile is one balancing authority. The rail marks its latest-published carbon intensity against the one gate every meter uses: 100 grams of CO₂ per kilowatt-hour, the same line the standard publishes. A gate reads open or shut from the grid data alone. Where a meter is installed the tile says so, and the certification calls live in that meter's cockpit; an open gate elsewhere certifies nothing, because nobody measured any compute there. The feed is delayed, and each tile carries its own measured delay rather than a promise of real time.
How often each gate opens
The board shows one hour; this table shows the years. For every authority, the share of published hours whose grid cleared the gate, computed from a seven-year backfill of the same federal feed. On most grids the honest number starts with a zero. On the hydro columns of the Pacific Northwest it does not, which is why the next meters are going there.
| 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 | 2026 | ||
|---|---|---|---|---|---|---|---|---|---|
| BPAT | 93.4% | 99.5% | 95.7% | 95.9% | 93.9% | 97.7% | 97.6% | 100.0% | 100.0% |
| CHPD | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% |
| DOPD | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% |
| GCPD | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% |
| GWA | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% |
| SCL | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% |
| SEPA | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% |
| SPA | 12.1% | 11.5% | 12.8% | 14.1% | 67.1% | 100.0% | 100.0% | 100.0% | 100.0% |
| TPWR | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% |
| WAUW | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% |
| YAD | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% |
| AVRN | 83.6% | 28.3% | 24.8% | 37.0% | 29.1% | 12.7% | 25.8% | 35.4% | 35.4% |
| AVA | 11.3% | 16.6% | 4.9% | 13.5% | 4.9% | 11.0% | 9.4% | 32.4% | 32.4% |
| CISO* | 9.1% | 0.9% | 0.3% | 1.6% | 0.4% | 0.3% | 0.1% | 29.5% | 29.5% |
| IPCO | 38.1% | 34.4% | 10.1% | 24.0% | 25.7% | 19.1% | 15.8% | 26.4% | 26.4% |
| PNM | 0.0% | 0.0% | 0.0% | 0.0% | 1.4% | 6.0% | 10.1% | 17.7% | 17.7% |
| PACW | 0.0% | 0.0% | 0.0% | 30.1% | 81.3% | 77.9% | 52.2% | 10.2% | 10.2% |
| PGE | 11.4% | 15.7% | 1.4% | 12.1% | 2.7% | 4.8% | 0.0% | 8.7% | 8.7% |
| WALC | 14.2% | 6.7% | 5.7% | 4.1% | 4.1% | 5.3% | 7.8% | 8.4% | 8.4% |
| SRP | 0.9% | 0.0% | 0.8% | 1.3% | 0.5% | 2.9% | 3.0% | 5.2% | 5.2% |
| PSEI | 0.0% | 0.6% | 2.8% | 1.8% | 0.2% | 0.4% | 2.7% | 2.0% | 2.0% |
| BANC | 4.8% | 0.3% | 0.0% | 0.0% | 3.5% | 0.3% | 0.0% | 1.6% | 1.6% |
| CPLE | 0.0% | 0.8% | 0.7% | 0.3% | 1.0% | 0.2% | 0.3% | 0.4% | 0.4% |
| AECI | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| AZPS | 1.7% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| CPLW | 0.0% | 0.3% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| DEAA | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| DUK | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| EPE | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| ERCO | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| FMPP | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| FPC | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| FPL | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| GRID | 0.0% | 0.0% | 0.3% | 0.0% | 0.0% | 0.1% | 0.0% | 0.0% | 0.0% |
| GVL | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| HST | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| IID | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.3% | 0.0% | 0.0% | 0.0% |
| ISNE | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| JEA | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| LDWP | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| LGEE | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| MISO | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| NEVP | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| NWMT | 0.1% | 0.1% | 0.0% | 0.0% | 0.0% | 3.0% | 1.5% | 0.0% | 0.0% |
| NYIS | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| PACE | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| PJM | 0.0% | 0.0% | 0.0% | 0.1% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| PSCO | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| SC | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| SCEG | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| SEC | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| SIKE | · | · | · | · | · | · | 0.0% | 0.0% | 0.0% |
| SOCO | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| SWPP | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| TAL | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| TEC | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| TEPC | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| TIDC | 0.7% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| TVA | 0.1% | 0.3% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| WACM | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
| AEC | 0.0% | 0.0% | 0.0% | · | · | · | · | · | n/a |
| EEI | 0.0% | 0.0% | · | · | · | · | · | · | n/a |
| GLHB | · | 0.0% | 0.0% | 0.0% | · | · | · | · | n/a |
| GRIF | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | · | · | · | n/a |
| HGMA | · | · | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% | · | n/a |
| NSB | · | · | · | · | · | · | · | · | n/a |
| WWA | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | · | n/a |
Share of published hours whose generation-mix carbon intensity was at or under 100 gCO2/kWh (the clean gate alone; not the meters' full-stack rarity).
* Demand is net of rooftop solar (EIA-930): midday figures understate total solar. Footprint: CAISO balancing authority.
Source: U.S. Energy Information Administration, Form EIA-930 (Hourly Electric Grid Monitor), public domain; seven-year hourly backfill, vintages and revisions as published; computed 2026-07-06T21:12:12Z.
Method
Data: EIA-930 Hourly Electric Grid Monitor (U.S. Energy Information Administration, public domain), served on its published lag and labeled with the measured delay. Carbon intensity is computed from each authority's reported fuel mix with the factor table published in the standard; the meters run the same computation on the same table. Water intensity is consumption, never withdrawal, on a cooling-typed regional basis with the basis named on every figure. Historical rates count the carbon gate alone; a meter's certification additionally requires the water gate and a measured job, so the table is an upper bound on what any grid could certify.
How the meter makes the call
An Oracle reads the live grid mix and facility state and computes the window's carbon intensity and water figures. The grid feed is delayed, never real-time, and labeled as such.
A Matchmaker applies both gates: the clean-energy gate (≤100 gCO₂/kWh) and the water gate (facility WUE strictly under 1.8 L/kWh, the labeled interim default where no basin prescription exists yet). When a window fails either, it holds. Holding through dirty and peak windows is the restraint the meter certifies, so the everyday state is a hold, not a mint.
A Provenance Stamper measures the job, aggregates every execution window into a composite record, and seals it with a sha256 hash over a fixed canonical form. Every new record carries the prescription in force at mint (its vintage). An optional signature is additive to the hash.
On the rare window that clears every gate, one credit is minted per certified GPU-hour, referencing the record. There is no volume counter and no celebration.
What is real, what is modeled, what is deferred
The honesty legend is first-class, not fine print. Every surface on this site and in the app carries these tags, and they never blur.
Measured or computed from live data now.
Derived from published figures, labeled as modeled.
Specified and severable, not built yet.
REAL the live delayed grid carbon signal; the indirect water computed from grid factors against the live mix; the telemetry energy basis; the gate, restraint, and mint logic; the provenance and sha256 hash with its in-hash vintage; the ledger; and the offline-verifiable bundle.
MODELED direct and cooling water via published water usage effectiveness benchmarks; facility profiles; and the simulated demo workloads the public meter runs, each labeled.
DEFERRED tradeability and any exchange (ceded by design); real facility water metering, which arrives with a pilot; the retired-GC reference (the granular_cert_ref field is null on every certificate today); and signing-key custody.
The rare hour
The hour that clears every gate mints one credit per certified GPU-hour, referencing the record and carrying the prescription in force at mint. The clean leg of that grade rides machinery the energy world already keeps; the water leg is the one Aquadrio originates, and it is what the credit attests. Hardware type and form factor are mandatory attributes, no cross-hardware equivalence is asserted, and water never changes the count; it only decides whether a clean window may mint.
The credit is not for sale.
Aquadrio mints to record the embedded cost and to prove the rarity. It does not offer credits to anyone, and the body never trades what it issues. Any future offering would follow legal review; until then the answer is no.