space-rha.comShort courseChapter 1 · The natural space environment

Chapter 1 · Lesson 6 of 6 · about 12 minutes

Writing the environment input package

Everything in this chapter ends up in one document: the set of spectra, fluences and dose-depth curves that the parts engineers, the system engineers and the test planners will use for the rest of the project. If that document is reproducible, labeled and complete, every later argument about a part can be traced back to it. If it is a list of numbers without models, versions, confidence levels and geometry, the arguments cannot be settled. This lesson lists what the document has to contain and fills it in for one mission.

What you need to know

An environment specification for radiation hardness assurance generally contains six things.

  1. The mission definition the numbers were computed for: orbit elements or trajectory, launch date and duration, and any phases (transfer, extended mission) treated separately.
  2. The models, with versions and settings: trapped-particle models and their solar-activity condition, the GCR model and modulation, the solar-event model with its confidence level and number of active years, the geomagnetic condition and cutoff method, the transport code and its geometry.
  3. The spectra themselves, as tables rather than single numbers: integral and differential proton and electron spectra, the heavy-ion LET spectra behind the reference shielding, and the solar-event spectra for both the cumulative and the design-event cases.
  4. The derived quantities at the reference shielding: dose-depth curves by particle, displacement-damage equivalent fluences, and the LET spectra, each with its geometry stated.
  5. The cases: which numbers apply to cumulative effects, which to event effects, which to the critical windows, and which solar-cycle phase each assumes.
  6. The limits: what the models do not cover (storm-time cutoff, model uncertainty, secondary particles, internal charging) and who owns the next step for each.

The reference shielding is a convention, usually 100 mil of aluminum in a solid sphere, chosen so that environments can be compared. The part-level requirements that follow replace it with the actual geometry (lesson 5 and chapter 2) and add the project's margin policy (chapter 3). The specification should say so explicitly, so that nobody downstream mistakes the reference numbers for the design numbers.

Worked example: the specification sheet for the 800 km mission

The question. Fill in the environment specification for the mission used throughout this chapter: sun-synchronous, 800 km, 98.6°, five years, launch 2028. The entries below use the runs already made for lessons 1 to 5. Entries that those runs do not cover are marked as open, which a real sheet should also do instead of filling them from memory.

ItemValueSource and settings
Orbit and duration800 km circular, 98.6°, 5 years from 2028Project; 7-day representative trajectory in SPENVIS; launch at the declining phase of cycle 25, so both solar-activity conditions are included
Trapped protons> 10 MeV: 168 cm⁻² s⁻¹; > 100 MeV: 62.8 cm⁻² s⁻¹ (orbit averages; full spectrum attached)AP-8 MIN, SPENVIS 4.6.14; AP-8 MAX run and AP9 percentile run open
Trapped electrons> 1 MeV: 1.7 × 10¹¹ cm⁻² per year (full spectrum attached)AE-8 MAX through the Space-RHA hardware panel; AE9 percentile run open
Galactic cosmic raysFe > 100 MeV/n: 0.368 m⁻² sr⁻¹ s⁻¹; LET spectrum behind 100 mil Al attached (600 km case)ISO 15390 at solar minimum with quiet cutoff (SPENVIS); CREME96 Space-RHA case at 600 km, 97.8°; rerun of CREME96 at 800 km, 98.6°, solar minimum, open
Solar protons, cumulative> 10 MeV: 7.6 × 10⁹ cm⁻²; > 100 MeV: 1.6 × 10⁸ cm⁻² at the orbit for one active year, 95 %ESP-PSYCHIC, SPENVIS 4.6.14, quiet transmission 0.21; rerun for the mission's number of active years open; stormy-magnetosphere case open
Solar protons and ions, design eventPSYCHIC worst event, 95 %, free space, with heavy-ion spectrum, then orbit transmissionLesson 3 table (GEO run, shadow removed); CREME96 worst-week, worst-day and peak-5-minute flux profiles for rate questions, open
Dose behind 100 mil Al, trapped2,580 rad(Si) per year (2,130 electrons and bremsstrahlung, 455 protons); 12,900 rad(Si) in 5 yearsSHIELDOSE-2 through IRBEM, center of a solid aluminum sphere, AE-8 MAX and AP-8 MAX; dose-depth curve 0.1 to 10 g/cm² attached (figure 5.1)
Dose behind 100 mil Al, solar505 rad(Si) per August 1972 event through the quiet cutoffReference only; the ESP active-year spectrum through SHIELDOSE-2 is the requirement basis, open
Displacement damageopenProton and electron spectra behind the reference shielding folded with NIEL (site NIEL tool); 10 MeV proton equivalent fluence to be stated
CasesCumulative: 5 years trapped plus ESP active-year term. Event: PSYCHIC 95 % event and CREME96 profiles. Windows: anomaly-pass peak flux (lesson 4) and event ratesLessons 3 and 4
LimitsStatic AP-8/AE-8 maps at 1960s and 1970s epochs; quiet cutoff; no secondaries; no internal charging assessment; sphere geometryOwners and next steps to be assigned

Steps.

  1. Attach the tables, not only the integral numbers. Each single value in the sheet is a summary of a spectrum, and the downstream analyses use the spectrum.
  2. Name the version of every model and tool, and the date of the run. The same orbit in AP9 at the 50th and 95th percentiles, or in a different SPENVIS release, gives different numbers, and a reviewer must be able to reproduce yours.
  3. Separate the cumulative case from the event case and from the windows, as lesson 4 did, so that no one verifies an event-rate requirement against a mission average.
  4. Mark what is open. The sheet above has six open items. A sheet with none is usually hiding assumptions.

Result. A specification that another engineer can rerun, that names which population drives which effect on this orbit (electrons for dose, anomaly and solar protons for displacement damage and proton SEE, galactic and solar heavy ions for heavy-ion SEE), and that hands the parts engineer the spectra rather than a single number per effect.

Going deeper

AE9/AP9 and percentiles

The AE9/AP9 models replace the static AE-8/AP-8 maps with a statistical description built on more recent data, and they report fluxes at a chosen percentile (50th for a mean mission, 95th for a conservative one) and as Monte Carlo scenarios that preserve the time correlation of the environment. For a requirement the percentile has to be stated, and for history-dependent effects the scenario runs are generally preferable to per-bin percentiles, which are not spectra any mission would actually see. Ginet et al. (2013) and Johnston et al. (2015) describe the models.

Confidence is not margin

A 95 % confidence level on a solar-proton fluence is a statement about the environment model's distribution. A radiation design margin of two on a dose is a project policy that covers model error, geometry uncertainty, part variability and test applicability together. The two are different things and should not be multiplied together as if they were the same. Chapter 3 treats the margin policy and the uncertainty budget that supports it.

Standards that define the package

ECSS-E-ST-10-04C specifies the models and conventions for European projects (AE-8/AP-8 or AE9/AP9 as agreed, ISO 15390 for cosmic rays, ESP-PSYCHIC for solar particles, 100 mil aluminum sphere reference). NASA/TM-20210018053 gives the equivalent guidance for NASA avionics, with the RHA process around it. SPENVIS and OMERE implement the models with documented settings; CREME96 remains the reference for the LET spectra and the solar-event flux profiles.

The common mistake

The common mistake is to issue the sheet once and never reopen it. A launch slip of two years moves the mission across a solar-cycle phase. An orbit raise of 200 km can double the trapped-proton flux. An extended mission doubles the cumulative terms. Each of those reopens the specification, and with it the part-level requirements that were derived from it. Keep the sheet under configuration control with the rest of the requirements.

Takeaway

The environment package is a set of reproducible spectra with their models, versions, confidence levels, geometry and cases stated, plus an explicit list of what is still open. A single number per effect is not a specification.

Check yourself

The sheet gives the solar-proton fluence for one active year. The mission is five years from 2028. What is the right next step?

Rerun ESP for the actual number of active years in the mission (the declining phase of cycle 25 and the rise of cycle 26 both fall inside it; the conventional active-year window is two years before maximum to four years after), at the project confidence level, and attach the full spectrum. Multiplying the one-year value by the number of active years is not equivalent because the model's cumulative percentiles are not linear in time.

The sheet gives the solar-proton fluence at 95 % confidence, and the project margin policy is a factor of two on dose. A colleague argues that the confidence level already is margin, so the factor of two can be dropped for the solar term. What is wrong with that?

The confidence level is a statement about the environment model's distribution: in the model, 95 % of such active years fall below the quoted fluence (lesson 3). The design margin is a project policy that covers model error, geometry uncertainty, part variability and test applicability together. They are different things, which is why the lesson warns against multiplying them together as if they were the same, and for the same reason neither stands in for the other. Chapter 3 treats the margin policy and the uncertainty budget that supports it.

A systems engineer asks why the sheet must attach full spectra when the parts engineers only need the fluence above 10 MeV. What is the reason?

Each single value in the sheet is a summary of a spectrum, and the downstream analyses use the spectrum: dose and displacement damage depend on its shape (lesson 3), the dose behind a given thickness depends on how much of the spectrum is soft (in lesson 5 the solar-event dose fell by a factor of 2.7 between 1 and 3 g/cm² while the trapped-proton dose fell by 1.6), and the displacement damage equivalent fluence is the spectrum folded with NIEL. Two environments with the same integral above 10 MeV can therefore give different doses at the part. The spectrum is also what a reviewer needs in order to reproduce the run.

A contractor delivers an environment sheet with every item filled in and no open items, including a displacement-damage fluence and an AP9 proton flux. Why does the lesson treat that as a warning sign, and what would you ask for?

A sheet with no open items is usually hiding assumptions; the sheet in this lesson, built from the chapter's own runs, has six. Ask for the model versions, run dates and settings behind each entry (the AP9 percentile, the solar-activity condition, the cutoff method, the transport code and its geometry), for the spectra behind each number, for the case each number serves (cumulative, event or window), and for the limits list with an owner for each item. The test is reproducibility: another engineer must be able to rerun the sheet and get the same numbers.

Two years after the sheet was issued, the launch slips by two years and the orbit is raised by 200 km. The parts list has already been qualified against the original sheet. Does the sheet need to be reopened?

Generally yes, and the part-level requirements with it. A two-year slip moves the mission across a solar-cycle phase, which changes the solar-event, cosmic-ray and low-altitude trapped-proton assumptions in different directions (lesson 3). A 200 km orbit raise can double the trapped-proton flux, which changes the displacement-damage and proton SEE terms (lesson 2). The qualification was against the original cases, so each requirement derived from the sheet has to be rechecked against the new ones, which is why the sheet belongs under configuration control with the rest of the requirements.

Further reading

  • ECSS-E-ST-10-04C Rev. 1, "Space engineering: space environment," 2020.
  • NASA, "Avionics Radiation Hardness Assurance (RHA) Guidelines," NASA/TM-20210018053, 2021. NTRS 20210018053.
  • K. A. LaBel et al., "Emerging radiation hardness assurance (RHA) issues: a NASA approach for space flight programs," IEEE Trans. Nucl. Sci., vol. 45, no. 6, pp. 2727-2736, 1998. doi:10.1109/23.736521.
  • G. P. Ginet et al., "AE9, AP9 and SPM: new models for specifying the trapped energetic particle and space plasma environment," Space Sci. Rev., vol. 179, pp. 579-615, 2013.
  • W. R. Johnston et al., "Recent updates to the AE9/AP9/SPM radiation belt and space plasma specification model," IEEE Trans. Nucl. Sci., vol. 62, no. 6, pp. 2760-2766, 2015. doi:10.1109/TNS.2015.2476470.
  • D. Heynderickx, B. Quaghebeur, E. Speelman and E. Daly, "ESA's Space Environment Information System (SPENVIS): a WWW interface to models of the space environment and its effects," AIAA 2000-0371, 2000; spenvis.oma.be.
  • M. A. Xapsos, G. P. Summers, J. L. Barth, E. G. Stassinopoulos and E. A. Burke, "Probability model for cumulative solar proton event fluences," IEEE Trans. Nucl. Sci., vol. 47, no. 3, pp. 486-490, 2000.
  • M. A. Xapsos, C. Stauffer, T. Jordan, J. L. Barth and R. A. Mewaldt, "Model for cumulative solar heavy ion energy and linear energy transfer spectra," IEEE Trans. Nucl. Sci., vol. 54, no. 6, pp. 1985-1989, 2007.