Chapter 1 · Lesson 1 of 6 · about 12 minutes
Three populations, three kinds of damage
Nearly every radiation requirement you write traces back to three particle populations: particles trapped in the Earth's magnetic field, particles thrown out by the Sun, and galactic cosmic rays. Each has its own energy spectrum, its own behavior in time and its own region of space. Which of them matters for a given part depends on the orbit, the shielding and the device, so the first step is to name the population before anyone names a test.
What you need to know
The near-Earth environment is generally described in terms of three sources.
- Trapped particles. The geomagnetic field holds energetic protons and electrons on closed drift shells. The inner zone is dominated by protons with energies up to several hundred MeV; the outer zone is dominated by electrons up to a few MeV. The populations overlap in space, and the outer electron belt can change by orders of magnitude within days. Trapped protons are the main source of displacement damage and proton-induced single-event effects in low Earth orbit, and trapped electrons generally set the total ionizing dose at higher altitudes.
- Solar energetic particles. Flares and coronal mass ejections accelerate protons and heavier ions. Events are irregular, last hours to days, and vary in size by orders of magnitude. The proton fluence of one large event can rival or exceed a year of trapped protons at some orbits (lesson 4), and solar heavy ions add single-event effects during the event.
- Galactic cosmic rays. A persistent, nearly isotropic background of fully ionized nuclei from hydrogen to uranium, with a broad energy spectrum peaking near 1 GeV per nucleon. Their flux is small, but a heavy ion such as iron deposits enough charge along its track to upset or destroy devices that nothing else in the environment can reach. Their energies are also too high for practical shielding to stop.
Three quantities link these populations to what happens in a device. Total ionizing dose (TID) is the ionizing energy absorbed per unit mass, in rad or gray, and comes mainly from trapped electrons, trapped protons and solar protons. Displacement damage dose (DDD) counts the energy that goes into knocking atoms out of the lattice, and comes mainly from protons and, in some orbits, electrons. Single-event effects (SEE) are caused by the charge from one particle track, and come from heavy ions (galactic and solar) and from protons through nuclear reactions and, in some modern devices, direct ionization.
| Population | Where | When | Main effects on electronics |
|---|---|---|---|
| Trapped protons | Inner zone, below about 2 Earth radii; the South Atlantic Anomaly for low orbits | Every pass through the region; slow secular change | DDD, proton SEE, part of TID |
| Trapped electrons | Outer zone, strongest near 3 to 5 Earth radii; horns reach down to polar LEO | Every pass; large storm-time variation | TID, internal charging, part of DDD |
| Solar protons and ions | Everywhere outside the geomagnetic cutoff; polar caps and high orbits | Irregular events, days long, more frequent near solar maximum | TID step, DDD, SEE during the event |
| Galactic cosmic rays | Everywhere, reduced inside the magnetosphere at low latitudes | Continuous; larger at solar minimum | Heavy-ion SEE |
The rest of this chapter looks at these populations one factor at a time: where you fly (lesson 2), the solar cycle and solar events (lesson 3), the time windows that matter (lesson 4), what shielding changes (lesson 5), and how to write all of it down as an input package (lesson 6).
Worked example: a one-year particle budget for the ISS orbit
The question. A camera electronics box will fly for one year at 400 km and 51.6°. Which population delivers the most particles, and does the answer change with the threshold?
Inputs. Orbit-averaged model outputs for that orbit, all from one SPENVIS project (runs of 15 October 2026): AP-8 MIN trapped protons, the ISO 15390 galactic cosmic-ray model at solar minimum with the quiet-time geomagnetic cutoff, and the ESP-PSYCHIC solar proton model at 95 % confidence for one active year with the quiet-time transmission of the orbit. The trapped-electron figure comes from the AE-8 MAX run behind the site's environment view (420 km, same inclination).
| Quantity | Model output | Per cm² per year |
|---|---|---|
| Trapped protons above 10 MeV | 17.6 cm⁻² s⁻¹ (AP-8 MIN) | 5.5 × 10⁸ |
| Trapped protons above 100 MeV | 5.70 cm⁻² s⁻¹ (AP-8 MIN) | 1.8 × 10⁸ |
| Trapped electrons above 1 MeV | AE-8 MAX, 420 km | 2.5 × 10¹⁰ |
| GCR hydrogen above 100 MeV/n | 446 m⁻² sr⁻¹ s⁻¹ (ISO 15390) | 1.8 × 10⁷ |
| GCR iron above 100 MeV/n | 0.210 m⁻² sr⁻¹ s⁻¹ (ISO 15390) | 8.3 × 10³ |
| Solar protons above 10 MeV, 95 %, quiet field | transmission 1 × 10⁻⁴ of free space | 6.5 × 10⁶ |
| Solar protons above 10 MeV, 95 %, stormy field | SPENVIS stormy-magnetosphere option | 2.6 × 10⁷ |
Steps.
- Convert flux to fluence. One year is 3.156 × 10⁷ s, so 17.6 cm⁻² s⁻¹ becomes 5.5 × 10⁸ cm⁻² per year.
- Convert the cosmic-ray unit. The GCR model reports flux per m² per steradian per second. For an isotropic field the omnidirectional flux through a sphere is 4π times that, and 1 m² is 10⁴ cm², so multiply by 4π × 10⁻⁴ = 1.257 × 10⁻³ to get cm⁻² s⁻¹, then by the seconds in a year.
- Compare at a common threshold. Trapped protons above 10 MeV outnumber the solar protons above 10 MeV by a factor of 86 at this orbit and confidence level, and outnumber galactic hydrogen above 100 MeV/n by a factor of 31.
Result. At the ISS orbit, trapped particles deliver nearly all of the fluence: about 2.5 × 10¹⁰ electrons above 1 MeV and 5.5 × 10⁸ protons above 10 MeV per cm² per year. The same solar-proton fluence would be 3.6 × 10¹⁰ cm⁻² outside the magnetosphere; the geomagnetic field passes only about one part in ten thousand of it to this orbit in quiet conditions, and about four times more in the stormy case. Galactic iron contributes only 8.3 × 10³ ions per cm² per year, yet it is the population that sets the heavy-ion single-event rate, because an iron ion deposits roughly 700 times the charge of a proton at the same speed along the same path (LET scales with the square of the charge, and 26² = 676).
What to carry forward. The ranking by count (electrons, protons, solar protons, galactic ions) differs from the ranking by effect. At this orbit, TID comes from the electrons and protons. Proton-induced SEE and displacement damage come from the trapped protons in the anomaly. Heavy-ion SEE comes almost entirely from the galactic ions. One "radiation level" cannot describe all three.
Going deeper
Why trapped particles stay trapped
A charged particle in a slowly varying magnetic field performs three periodic motions with widely separated time scales: gyration about a field line (microseconds to milliseconds), bounce between mirror points along the field line (tenths of a second to seconds), and drift around the Earth (minutes to hours). Each motion has an adiabatic invariant that is conserved while the field changes slowly compared with that motion. The first invariant, the magnetic moment p⊥²/(2mB), gives the mirror condition sin²α_eq = B_eq/B_mirror. Particles whose mirror points lie inside the atmosphere are lost, which empties the loss cone. The third invariant is broken by storm-time field changes on the drift time scale, which is one of the mechanisms that move electrons into and out of the outer zone. The gyroradius of a 100 MeV proton at 1.2 Earth radii is tens of kilometers, so spatial gradients in the inner zone are not always slow in the required sense. For that reason the inner-zone proton models are empirical maps rather than solutions of these equations.
Galactic cosmic-ray composition
By number, galactic cosmic rays are roughly 87 % protons, 12 % helium nuclei and about 1 % heavier nuclei (Bourdarie and Xapsos 2008 give these approximate fractions). Within the heavy fraction the abundances follow the solar-system pattern with two exceptions caused by spallation during propagation: the light elements Li, Be and B and the sub-iron group are strongly enhanced relative to solar abundances. Iron is the heaviest nucleus with an appreciable flux, and because linear energy transfer scales roughly with Z², iron sets the upper edge of the LET spectrum that most of the SEE flux occupies (lesson 2). Nuclei heavier than iron exist in the spectrum and in the CREME96 tables (to Z = 92), at fluxes several orders of magnitude lower.
Which solar-event population matters for which effect
Solar protons deliver the fluence that matters for dose and displacement damage during an event, and their nuclear reactions in silicon produce the recoils that cause proton-induced single-event effects. Solar heavy ions are far fewer, but in a large event their flux above a given LET can exceed the galactic background by orders of magnitude for hours to days. A device with a low LET threshold can therefore see most of its mission heavy-ion events during a handful of solar events. The 95 % worst-event spectrum in lesson 3 gives the numbers for both.
The common mistake
The common mistake is to treat one number as the environment. A statement such as "the orbit sees 10 krad" names one effect of one population behind one shielding assumption. It says nothing about the proton fluence that will degrade an optocoupler, or the iron flux that will upset a memory. Ask which population and which effect before accepting any single figure, and keep the three populations in separate columns in every environment table you produce.
Takeaway
Name the population and the effect before naming a level: trapped, solar and galactic particles reach a part at different times, through different shielding, and damage it in different ways.
Check yourself
At the ISS orbit, trapped protons above 10 MeV outnumber galactic iron above 100 MeV/n by about 70,000 to one. Why does the iron still need its own line in the requirement?
Because single-event effects depend on the charge deposited by one track, not on the number of tracks. An iron ion deposits roughly 700 times the charge of a proton of the same velocity (LET scales roughly with Z², and 26² = 676), so it can reach the critical charge of devices that protons cannot upset directly. The proton population is handled by a separate proton SEE assessment, usually through nuclear-reaction recoils.
At the ISS orbit the one-year solar-proton fluence above 10 MeV at 95 % confidence is 6.5 × 10⁶ per cm², a factor of 86 below the trapped protons. A colleague proposes dropping solar protons from the ISS specification altogether. What about that number should give pause before agreeing?
The figure is the free-space fluence of 3.6 × 10¹⁰ per cm² passed through the quiet-time geomagnetic cutoff, which admits only about one part in ten thousand of it to this orbit. The SPENVIS stormy-magnetosphere option gives 2.6 × 10⁷ per cm², about four times more, and the transmission depends on the energy and on the geomagnetic condition assumed, so it is a model setting rather than a margin (lesson 2). The solar term is generally small at this orbit, but it belongs in the sheet with its confidence level, its number of active years and its geomagnetic condition stated, so that a reviewer can see why it is small rather than find it missing.
A galactic cosmic-ray model reports 446 protons per m² per steradian per second above 100 MeV/n. An engineer multiplies by 10⁻⁴ to convert the area to cm² and by 3.156 × 10⁷ s to get the fluence for a year. What has been left out, and why does it belong in the conversion?
The factor of 4π. The model gives a directional flux per steradian, and for an isotropic field the omnidirectional flux through a sphere is 4π times that, so the full conversion to cm⁻² s⁻¹ is 4π × 10⁻⁴ = 1.257 × 10⁻³, which gives 1.8 × 10⁷ per cm² per year at the ISS orbit. Without it the fluence is understated by 4π. Model outputs come as integral or differential and as directional or omnidirectional, and a quantity is only usable when all four attributes are labeled (lesson 4).
The pitfall warns that "the orbit sees 10 krad" names one effect of one population behind one shielding assumption. Requirements are being written for an optocoupler and for a memory on the same board in a low Earth orbit. Which population and which quantity does each need, and why does the dose figure serve neither?
The optocoupler degrades by displacement damage, which in low Earth orbit comes mainly from the trapped protons of the anomaly and from solar protons, so it needs the proton spectrum and the displacement damage dose derived from it, not the ionizing dose. The memory upsets by single-event effects, which need the heavy-ion LET spectrum of the galactic and solar ions, plus a separate proton SEE assessment through nuclear-reaction recoils. The dose figure comes mainly from trapped electrons and protons, and the three effects generally do not track one another from one orbit to the next, so one level cannot stand in for the other two.
A GEO project takes its trapped-electron flux from one week of measurements by a spacecraft in a similar orbit and calls that the environment. What in this lesson argues against it?
The outer electron belt can change by orders of magnitude within days, with large storm-time variation on every pass, so one week is a sample of the belt in one state rather than a description of what the mission will meet. The trapped-particle models exist to describe the long-term population, and the newer ones report it at a stated percentile (lesson 6), which is what a requirement needs. A short measurement is generally useful as a cross-check on the model, not as a replacement for it.
Further reading
- S. Bourdarie and M. Xapsos, "The near-Earth space radiation environment," IEEE Trans. Nucl. Sci., vol. 55, no. 4, pp. 1810-1832, 2008. NTRS 20080039622.
- J. L. Barth, C. S. Dyer and E. G. Stassinopoulos, "Space, atmospheric, and terrestrial radiation environments," IEEE Trans. Nucl. Sci., vol. 50, no. 3, pp. 466-482, 2003.
- NASA, "Avionics Radiation Hardness Assurance (RHA) Guidelines," NASA/TM-20210018053, 2021. NTRS 20210018053.
- D. M. Sawyer and J. I. Vette, "AP-8 trapped proton environment for solar maximum and solar minimum," NSSDC/WDC-A-R&S 76-06, 1976.
- J. I. Vette, "The AE-8 trapped electron model environment," NSSDC/WDC-A-R&S 91-24, 1991.
- ISO 15390:2004, "Space environment (natural and artificial): galactic cosmic ray model."
- Space-RHA, "Space radiation environment in 3D," space-rha.com/space-radiation-environment (the hardware panel supplies the dose figures used here).