From low Earth orbit to the slot region, from the brutal heart of the Van Allen belts to the quiet of deep space — the radiation environment changes everywhere you go. This is the journey, and these are the consequences.
Below about 1,000 km, Earth's atmosphere — even at its barest — and our magnetic field shield us from most of the cosmos. The International Space Station orbits here at ~408 km. So do most Earth-observation satellites, communications constellations, and crewed missions.
The dominant threat at this altitude is atomic oxygen, not radiation. Dose rates are modest. But there's one exception that everyone working in LEO must respect: the South Atlantic Anomaly, where the inner proton belt dips down to brush against us.
Behind 100 mil (≈2.5 mm) of aluminum, total ionizing dose at 400 km averages 1–3 krad/yr — well within commercial-grade parts capability. ISS hits the SAA 6–8 times per day during ~5% of its time on orbit, but those passes account for roughly half of crew absorbed dose. By dose equivalent (with biological weighting), GCR dominates because heavy ions are more damaging per unit energy.
Earth's magnetic dipole is offset from its rotational axis — and tilted. The consequence: over the South Atlantic, the inner proton belt dips down to as low as 200 km. Satellites passing through experience a sudden, dramatic spike in particle flux.
The Hubble Space Telescope's instruments shut down through SAA passes. Astronauts on ISS sometimes see cosmic ray flashes with eyes closed. Single-event upsets on commercial satellites cluster here — you can predict them on a map.
The SAA is slowly migrating westward and weakening (the broader phenomenon of geomagnetic field weakening). It causes most LEO SEUs and forces sensitive instruments into "safe mode" during transits. The protons here are fast — high enough energy that thin aluminum doesn't help much.
Above ~1,000 km, you enter the inner Van Allen radiation belt, discovered by James Van Allen's instrument on Explorer 1 in 1958. This is a torus of high-energy trapped protons — some exceeding 400 MeV — held captive by Earth's magnetic field, bouncing pole to pole and drifting around the planet.
The peak sits near 3,000 km. Almost nothing operates here long-term. Spacecraft transit through quickly on their way to higher orbits. The protons are penetrating: aluminum shielding helps modestly, but going thicker fast hits diminishing returns.
Trapped protons follow magnetic field lines, bouncing between magnetic mirror points near the poles. They also drift around the planet — westward for protons, eastward for electrons. Penetrating energies mean shielding curves flatten beyond ~500 mil Al; designers turn to rad-hard parts instead.
Between the two belts sits a region of genuine respite: the slot region. Wave-particle interactions scatter trapped electrons out of this band, leaving a relatively benign corridor at around 10,000 km.
Some constellations exploit this — O3b mPOWER and SES's medium-earth-orbit fleets live here precisely because dose rates are an order of magnitude lower than the surrounding belts. The catch: the slot can fill in during severe geomagnetic storms, sometimes for weeks.
The "slot" is created by plasmaspheric hiss and chorus wave-particle interactions that scatter electrons into the loss cone. During severe geomagnetic storms the outer belt expands inward and the slot can disappear entirely. The March 1991 storm injected a transient third electron belt that persisted for over a year. Modern slot constellations design for storm-time excursions, not steady-state.
Most people assume "higher is safer." It isn't. Between 13,000 and 30,000 km sits the outer electron belt — a vast swarm of relativistic electrons, peaking around 16,000–20,000 km. Right where GPS lives.
GPS satellites at 20,200 km sit inside this belt every minute of every day. They're built like tanks — radiation-hardened parts, thick shielding, redundant electronics. The dose rate behind modest shielding is roughly 20–30 krad/year — cumulative over a 15-year design life, that's hundreds of krad. Electrons also cause deep dielectric charging: charge builds inside insulators until it arcs catastrophically.
Outer-belt electrons are highly variable — driven by solar wind, geomagnetic storms, and substorms. Flux can jump 1000× in hours. The "killer electron" events of October 2003 destroyed several spacecraft. MEO designers assume worst-case storm conditions plus large margins.
GEO sits at the outer edge of the outer belt. The environment is more benign than peak MEO — but not by enough to relax. The dominant threats here are surface charging from low-energy plasma during substorms, and solar particle events that arrive with little protection from Earth's field.
A geomagnetic storm at GEO can build kilovolts of differential charge across a satellite's surface in minutes. When that charge arcs to the chassis, electronics can latch up, sensors can blind, command-and-control can be lost. The Galaxy 15 anomaly in 2010 — "Zombiesat" — was a charging event.
GEO is the workhorse orbit for commercial comms and weather — its environment is well-characterized and predictable in the long run. Modern GEO satellites are designed for 15+ years on-orbit. The dominant risk shifts from total dose to single-event effects and charging anomalies.
A few times per solar cycle, a flare or coronal mass ejection accelerates protons to relativistic speeds and hurls them at the inner solar system. Inside the magnetosphere we're partially protected. Outside, you're exposed to years of dose in hours.
The August 1972 event happened between two Apollo missions; had astronauts been on the lunar surface, unsheltered, they would have received a potentially lethal dose. Carrington 1859 — the largest event in modern history — was inferred from ice cores and telegraph fires.
Carrington 1859 was the largest SPE in 450 years — ~4× the August 1972 event, the largest of the spacecraft era. Fluences reference >30 MeV protons. SPEs are episodic and largely unpredictable in detail. Modern crewed mission designs assume one worst-case SPE per mission and provide a storm shelter. Hardware designs use SPE peak-flux statistics from missions like GOES.
Beyond Earth's magnetosphere — past the Moon, on the way to Mars, anywhere in deep space — the dominant radiation source becomes galactic cosmic rays. These are atomic nuclei accelerated by ancient supernovae, traveling at near-light speed. Many are heavy ions: iron, oxygen, silicon.
Here's the cruel part: you can't shield against them. Thin aluminum makes things worse, not better — primary GCRs fragment into showers of secondaries. Effective shielding requires either tons of hydrogen-rich material or many meters of regolith. The Sun helps: when solar activity is high, the heliosphere puffs up and deflects GCR. So GCR is highest near solar minimum, exactly inverse to SPE risk.
GCR is the rate-limiter for human deep-space exploration. The NASA career limit is 600 mSv effective dose — a Mars round-trip approaches that. Heavy ions (the 1% with Z>2: carbon, oxygen, iron) produce dense ionization tracks that damage DNA in ways photons don't — biological effectiveness is high. Solar-cycle modulation comes from Chen et al. 2023 with AMS-02 measurements.
For most spacecraft, shielding means aluminum — it's structural, it's already there, it does double duty. The first 100 mil (2.5 mm) of aluminum cuts trapped-electron dose by roughly an order of magnitude. The next 100 mil cuts it by less than half again. Beyond ~500 mil, you've stopped helping.
The curve below shows dose-vs-thickness for different orbits. This is one of the most-used charts in spacecraft engineering — every design review includes one. The answer is rarely "more aluminum." It's rad-hard parts, careful layout, and design margin.
Five missions, five very different radiation environments — and the design decisions that follow from them.
Pick an altitude. Pick a solar cycle phase. Pick a shielding thickness. See what your spacecraft is up against — and what it needs to survive.