The next weather-satellite fleet was cut from six to four and $19.6 billion to $12 billion. It keeps its 2032 deadline by flying spare parts.

The finding, in one paragraph: every satellite image in every hurricane forecast comes from the GOES system, whose current spacecraft age out in the mid-2030s, and whose replacement — GeoXO — is being rebuilt mid-program around a spending cap rather than a requirement.▸ The program the Commerce Department's Inspector General audited in 2023 had established baselines: $19.6 billion, six spacecraft, first launch 2032, instruments chosen through a decade of formal user studies. The program that exists after the 2025 budget passback is four spacecraft under $12 billion, holding the same 2032 date by flying a spare previous-generation imager, minus two instruments cancelled after $852 million of development, minus the NASA procurement partnership that has bought American weather satellites since the 1970s. During hurricane season, with the forecast machinery under strain BlackLeaf has documented elsewhere, the question the documents pose is narrow: what does the cap buy, and what did it silently spend?

Five sources.▸ The audited baseline is the Commerce OIG's The GeoXO Program: Cost and Schedule Baselines Are Established... (OIG-23-028-A, September 2023 — four recommendations, $32.8 million flagged as funds for better use). The budget record is CRS's NOAA FY2026 appropriations brief (read directly in PDF), documenting the passback, the "skinny" request's GeoXO "rescope," and the Technical Supplement's continuity claim. The restructure's specifics — the cap, the four-satellite fleet, the cancelled BAE instruments, the spare-parts plan — are the SpaceNews and Payload records of NOAA's and OMB's decisions. The pushback is the Senate Commerce Committee's unanimous March 2026 advancement of the Weather Research and Forecast Innovation Reauthorization Act, directing imaging and sounding on all GeoXO satellites.

The arithmetic of the cancellations is the sunk-cost problem in miniature: the atmospheric-composition and ocean-color instruments were not proposals — they were contracts in execution, $852 million committed at BAE Systems, terminated as "climate" instruments under the budget language even though air-quality sounding feeds the same forecast models as everything else on the spacecraft.▸ The fixed-price conversion and the dissolution of the NASA joint office are the structural bets: that commercial-style procurement can build civil weather satellites cheaper than the NASA partnership that built every previous generation — a bet with no American track record, placed on the system whose failure mode is a blind spot over the Atlantic in September.

Holding 2032 is the restructure's proudest claim, and the documents show how it is held: the first GeoXO satellite will fly a spare imaging sensor from GOES-R-era storage and a lightning mapper assembled from existing parts.▸ The successor program's first spacecraft launches with its predecessor's instruments — continuity achieved by definition, advancement deferred to satellites three and four, under a fixed annual cap that makes every future problem a schedule problem.

Three document pairs.▸ The audit vs the passback. The OIG's 2023 finding was that baselines were established — its concerns were about optimizing the Central satellite and pricing performance gains, recommendations addressed to a $19.6 billion program. The passback answered a different question no audit asked: what fits under $500 million a year. No public analysis reconciles the two — the requirement-derived program and the cap-derived one have never been compared in a document with numbers. The budget vs itself. The Technical Supplement asserts the request "maintains continuity of major systems, including satellites, needed for weather forecasting" — while rescoping the fleet from six to four and cancelling instruments; CRS notes drily that "it is unknown what activities may fall under modernization efforts." Congress vs OMB, again. The Senate Commerce Committee's unanimous March 2026 bill would statutorily require advanced imaging and sounding on all GeoXO satellites — the same pattern BlackLeaf documented at NSF and Artemis: an executive cap, met by a congressional floor, with the program suspended between them.

The FY2027 appropriations cycle prices the standoff: the House and Senate CJS bills will either fund the four-satellite cap or the six-satellite statute-in-waiting.▸ The BAE terminations are final; restoring those measurements later means re-procuring from zero. GOES-19 and its siblings keep the margin honest — every year of GeoXO slip under the fixed cap spends the three-year buffer the 2032 date protects. And the Weather Act reauthorization, if it passes the floor as it left committee, converts the fleet's capabilities from a budget decision into a legal one.

Baseline figures are from the DOC OIG audit; restructure figures are from the trade record of NOAA's and OMB's decisions; budget language is quoted from the CRS brief, read directly. Design-life margin is approximate; GOES spacecraft frequently exceed design life, which is a hope, not a plan.