The question: beyond the magnetosphere, radiation is the constraint. What do we actually know about dose at the Moon, on the way to Mars, and at Mars — and what is still guesswork?
We found solar-particle events that produced order-of-magnitude dose responses at the Moon while Earth’s official catalogs stayed silent — missed by threshold (too small at Earth), by latency (arrived at the Moon first), and by curation (never entered the record). Each was confirmed by an independent spacecraft in lunar orbit. If you operate near the Moon, you need lunar-vicinity monitoring, not just Earth-side alerts.
Counting the same way across 17 years, the quiet-day radiation bursts that seeded those discoveries are overwhelmingly a current-cycle phenomenon — 44 of 60 burst-days live in 2023–2025, and 2012 (a loud year by Earth standards) has zero. Models tuned on past cycles will miss this Sun.
One structural change — letting storm hours and quiet hours carry separate weights — fixed a 18× storm-hour error and stayed fixed across the entire archive. The magnetotail’s effect on dose still behaves opposite to textbook expectation; that stays an open question, published, not hidden.
Solar-particle events suppress surface dose at Mars — the same direction in both measured years — and in one season we certified the effect as essentially zero, firmly. (A confident zero is real knowledge: it bounds the hazard.) The same solar events push dose up at the Moon: same Sun, opposite signs, ordered by shielding. One rule, planetary bodies as its coordinates.
Our residuals flagged them, and upstream plasma data from an orbiting spacecraft confirmed both, to the day — arrivals no official catalog lists.
A seasonal dose signal at Gale Crater appears in both measured years and has survived every explanation we tested — it is not calibration, not daily storm bookkeeping, not maskable. It waits on a dust-opacity data release or a Mars upstream monitor; our prediction for the ESCAPADE mission (~2028) is already staked in advance.
| gap | what closes it |
|---|---|
| Earth→Mars corridor: crew transit dose has essentially never been continuously measured — 4 one-way snapshots in 25 years | a small hosted dosimeter on any Mars-bound rideshare, hourly, one solar rotation; analysis pre-committed on our side |
| Lunar surface: the measurements exist (the LND instrument on the farside) but archive access is blocked | one data-access agreement — and it matters twice: 17 years of lunar-orbit data hold only 46 magnetotail-storm days, so a farside stream is the only way to close that cell |
LRO/CRaTER dose rates (UNH, full mission 2009–2026) · MSL/RAD surface dose (NASA PDS, two eras) · MAVEN SWIA/MAG upstream plasma · GOES proton/electron feeds (three instrument eras) · NASA OMNI · NASA DONKI arrival catalogs. Every fetch sha-pinned; the one gap in the archives themselves (an 8-month proton-coverage hole in 2020 between satellite generations) is documented in the register.
Every claim on this page comes from a pre-registered analysis: the hypothesis and its pass/fail rule were committed to a ledger before the data was touched, results were reproduced bit-for-bit, decoy signals ran as tripwires, and refused hypotheses stay on record. The complete gap register — data sources, refused hypotheses, and open staked predictions — is published alongside this atlas (how to read it).
Nervous Machine