Water Shield Radiation Evaluation (WSRE)

What this page is

This page defines a neutral evaluation class for water-based radiation shielding at habitat scale. It specifies what must be modeled, measured, and reported before a thick-water shield can be credited with a protection level in any human-rating argument. It is not a mission design, not a product announcement, and not a claim about the answer.

The thick-shield question

For short missions, shielding analysis can lean on thin-shield intuition: more mass, less dose, roughly exponentially. A habitat intended for decades of occupancy under career-dose limits (NASA‑STD‑3001 currently sets 600 mSv career effective dose) lives in a different regime:

• Fragmentation, not absorption. Galactic cosmic ray heavy ions do not stop in the first tens of g/cm² — they fragment into lighter, longer-ranged particles. Dose equivalent falls far more slowly than fluence.
• Secondary buildup. Neutrons, light fragments, gammas, and (at depth) pions and muons are produced inside the shield itself. Behind multi-meter water, the interior dose is increasingly carried by particles that did not exist outside it.
• Model divergence. High-energy nuclear cross sections carry real uncertainty, and independent transport codes diverge most at large areal densities — precisely where habitat-class shields operate. A single-code answer is not a verified answer.

Water is among the best bulk shielding materials per unit mass (high hydrogen content, favorable fragmentation), which is why it is the reference material here. The evaluation exists to convert that qualitative advantage into a quantified, defensible protection curve over the credible thickness range (roughly 1–5 m, ~100–500 g/cm²) — and to map how any point on that curve degrades under realistic geometry, penetrations, and off-nominal fill states.

Radiation environment envelope

The evaluation shall cover the complete free-space environment at the habitat's operating location (cislunar / deep space; no geomagnetic shielding credited, no trapped-belt population).

Galactic cosmic rays

All GCR cases are isotropic over 4π. Species: protons, helium, and all heavy ions Z = 3–28 individually (with C, N, O, Ne, Mg, Si, Fe explicitly resolved — iron is the single largest dose-equivalent contributor at depth), plus electrons and positrons as a boundary term.

Solar particle events

Each SPE case shall be evaluated as event-integrated fluence; the worst-case (1972-class and Carrington-class) shall additionally be evaluated with a time-dependent profile to produce peak hourly dose rates relevant to operational response and 30-day limits.

Species and energy coverage

Evaluation objectives

Functional requirements (requirements-style)

Note: quantitative thresholds (dose targets, agreement bounds, confidence levels) are architecture-dependent and are set by the program's human-rating requirements; representative values appear below.

Interior dosimetry locations

Graduated evaluation tiers

The campaign proceeds in tiers, each retiring risk for the next:

• Tier 1 — Deterministic scoping. HZETRN/OLTARIS-class parametric sweeps over the full areal-density range, environments, and additives. Fast, broad, establishes the shape of the dose-vs-thickness curve and the sensitivity ranking. Not creditable as verification on its own.
• Tier 2 — Cross-validated Monte Carlo. Full 3D geometry in ≥2 independent codes at selected points along the thickness range, with phantom dosimetry, degraded states, and the complete uncertainty budget. This tier produces the numbers any sizing decision stands on.
• Tier 3 — Beam anchoring. Accelerator campaigns on instrumented water targets at staged areal densities, including GCR-simulator mixed-field runs where available. Experiment–simulation comparison closes the loop on the code predictions at the depths that matter.

The terrestrial analog is standard practice: reactor and accelerator shielding is never certified on a single unvalidated code. The same discipline applies here; the physics is harder, not easier.

Sensitivity studies

Acceptance criterion classes

The evaluation shall report against the following criterion classes. Numerical thresholds are deliberately not asserted here: setting a target dose in advance of the evaluation would presume its conclusion. Thresholds derive from the program's human-rating requirements and current NASA‑STD‑3001 / ICRP guidance at the time of evaluation.

Data products

Usability requirement: outputs must be legible to program managers and human-rating reviewers, not only radiation physicists.

Ballpark cost (ROM)

Order-of-magnitude, planning-grade only: a Tier 1–2 simulation campaign of this scope is a low single-digit $M effort dominated by analyst time and compute; a Tier 3 accelerator validation campaign is likewise low single-digit $M, dominated by beam time, targets, and instrumentation. Either figure is small against the cost of committing habitat-scale structure to an unverified protection level — or of over-building shield mass the physics does not require.

Relationship to mission risk

Absence of WSRE-class evaluation before committing to a shield architecture constitutes a known program risk in both directions: an under-performing shield discovered after construction is unrecoverable, and an over-built shield wastes the single most expensive commodity in the architecture — delivered mass. Chronic GCR exposure is the binding constraint on permanent habitation; it deserves the same evidentiary standard as structure and life support.

WSRE-class evaluation reduces epistemic uncertainty; it does not eliminate radiobiological uncertainty in risk models, and it does not replace onboard dosimetry and operational monitoring.

Standardization potential

Standardizing WSRE-class geometry files, environment definitions, and output formats enables cross-comparable results between independent teams and cumulative refinement as codes and cross-section libraries improve. Over time this constitutes a shared verification baseline for water-shielded habitats generally — a benchmark class, not a one-off study.