Skip to content

Phase 61: Maritime, Acoustic, & EM Environment

Block 7 — Final Engine Hardening Status: Complete Tests: 71 new (8,344 total Python passing)

Overview

Wires three clusters of environmental subsystems whose outputs were computed but never consumed by the simulation loop: sea state effects on ship operations, underwater acoustic layers on sonar detection, and electromagnetic propagation on radar/comms/DEW.

Deliverables

Step 0: Infrastructure

  • stochastic_warfare/simulation/calibration.py — 3 new boolean fields: enable_sea_state_ops, enable_acoustic_layers, enable_em_propagation (all default False)
  • stochastic_warfare/simulation/scenario.py — UnderwaterAcousticsEngine, EMEnvironment (populates conditions_engine), and CarrierOpsEngine instantiated on SimulationContext; state persistence (get_state/set_state); EM environment wired to CommunicationsEngine
  • stochastic_warfare/simulation/engine.py — UnderwaterAcousticsEngine.update(dt) call in tick loop

61a: Sea State → Ship Operations (19 tests)

  • stochastic_warfare/simulation/battle.py (movement section):
  • Small craft speed penalty: −20% per Beaufort above 3 (displacement < 1000t or max_speed < 15)
  • Tidal current adjustment: effective_speed += current_speed × cos(current_dir − heading)
  • stochastic_warfare/simulation/battle.py (engagement section):
  • Wave period resonance: if |wave_period − hull_natural_period| < 10%, crew_skill × (1/1.5)
  • Swell direction: crew_skill × (1.0 − sin²(wave_dir − heading) × 0.5)

61b: Acoustic Layer → Sonar Detection (19 tests)

  • stochastic_warfare/simulation/battle.py (detection section):
  • Thermocline: target below, observer above → detection_range × 0.1 (~20 dB loss)
  • Surface duct: both in duct → × 3.0 (+10 dB gain); target below duct → × 0.06 (+15 dB loss)
  • Convergence zones at 55km intervals: CZ spike × 2.0; acoustic shadow between CZs × 0.05

61c: EM Propagation → Radar/Comms/DEW (16 tests)

  • stochastic_warfare/simulation/battle.py (detection section):
  • Radar horizon gate: 4/3 Earth model, antenna height derived from domain (ground ~10m, ship ~30m, aircraft = altitude)
  • EM ducting: maritime platforms with ducting → radar range extended (capped at 2.0×)
  • stochastic_warfare/simulation/battle.py (engagement section):
  • DEW humidity/precipitation_rate extracted from WeatherEngine and forwarded through route_engagement
  • stochastic_warfare/combat/engagement.py — humidity and precipitation_rate optional params added to route_engagement(), forwarded to execute_laser_engagement()
  • stochastic_warfare/c2/communications.py:
  • set_em_environment() method for EM environment injection
  • HF radio: reliability × hf_propagation_quality() (day ~0.3, night ~0.8)
  • VHF/UHF: beyond radio horizon → reliability × 0.1

Structural Verification (8 tests)

Source-level pattern checks confirming all wiring present across 4 source files.

Files Modified (6)

File Changes
stochastic_warfare/simulation/calibration.py 3 new boolean fields
stochastic_warfare/simulation/scenario.py 3 engine instantiations, state persistence, EM→comms wiring
stochastic_warfare/simulation/engine.py UnderwaterAcousticsEngine.update(dt)
stochastic_warfare/simulation/battle.py Sea state ops, acoustic layers, radar horizon, ducting, DEW params
stochastic_warfare/combat/engagement.py humidity/precipitation_rate forwarding
stochastic_warfare/c2/communications.py set_em_environment(), HF quality, radio horizon

New Test Files (5)

File Tests
tests/unit/test_phase_61_infra.py 9
tests/unit/test_phase_61a_sea_state_ops.py 19
tests/unit/test_phase_61b_acoustic_layers.py 19
tests/unit/test_phase_61c_em_propagation.py 16
tests/unit/test_phase_61_structural.py 8

Deferrals (Planned → Deferred)

  1. CarrierOpsEngine full battle loop wiring — instantiated on context but CAP management, sortie turnaround, recovery window enforcement require air sortie dispatch architecture. Structural prep only.
  2. Beaufort > 6 helicopter deck landing abort — no helicopter-carrier recovery tracking exists yet
  3. Beaufort > 7 carrier flight ops suspension — requires CarrierOpsEngine battle loop integration
  4. Landing craft 10% casualty risk at Beaufort > 5 — no landing craft type detection
  5. Sea spray/salt fog — maritime atmospheric obscurant at high Beaufort, needs ObscurantsEngine integration
  6. SHF/EHF rain attenuation for comms — comms engine doesn't track per-equipment frequency; approximate by comm_type categories later
  7. Ionospheric storm effects on HF — needs space weather events
  8. Hull natural frequency per ship class — hardcoded ~10s destroyers, ~12s carriers; should come from YAML data

Design Decisions

  1. 3 enable flags (not per-subsystem): Groups related effects. enable_sea_state_ops covers all Beaufort/tidal/wave. enable_acoustic_layers covers all sonar environment. enable_em_propagation covers radar horizon + comms + DEW.
  2. conditions_engine populated with EMEnvironment: Was always None on SimulationContext. Now SpaceEngine (which already passes em_environment=ctx.conditions_engine) gets a real EM environment, and battle.py/comms can use it for radar horizon and HF quality.
  3. CarrierOpsEngine structural only: Instantiate on context but defer full integration. CAP management needs air sortie dispatch that doesn't exist.
  4. Antenna height approximation: No antenna_height field on sensors. Derived from unit domain: ground ~10m, ship ~30m, aircraft = altitude.
  5. Safe attribute access throughout: All getattr(ctx, "field", None) — prevents SimpleNamespace mock breakage.
  6. DEW params through route_engagement: Added optional params to existing signature rather than modifying battle.py call to DEW engine directly. Follows established routing pattern.
  7. Rain detection factor confirmed already wired: Phase 52b wired _compute_rain_detection_factor() at battle.py:2318. Verified and skipped.

Lessons Learned

  • conditions_engine was a latent integration gap: Field existed on SimulationContext since early phases, SpaceEngine tried to use it, but nobody ever populated it. Phase 61 fixes this incidentally while wiring EM propagation.
  • Sonar sensor types differ from plan: Plan referenced HULL_MOUNTED, TOWED_ARRAY, SONOBUOY, DIPPING but actual SensorType enum uses ACTIVE_SONAR, PASSIVE_SONAR, PASSIVE_ACOUSTIC. Always verify enum values before coding.
  • Acoustic modifier math compounds multiplicatively: Thermocline (0.1) × shadow (0.05) = 0.005 — submarine below thermocline between CZ ranges is effectively invisible. This is physically correct but may surprise calibrators.

Postmortem

Scope

On target. 6 source files modified, 5 test files, 71 tests. Plan called for ~55 tests; delivered 71 (agents added extra edge cases). All 3 environmental clusters wired. CarrierOpsEngine structural-only is a documented deferral consistent with Phase 53/58 precedent.

Quality

High. No TODOs, FIXMEs, or hacks in modified source files. All effects gated by enable_*=False — zero behavioral change unless opted in. 8,344 tests pass with no regressions.

Integration

Fully wired for gated effects. UnderwaterAcousticsEngine on context + updated per tick. EMEnvironment populates conditions_engine (fixes SpaceEngine gap). CarrierOpsEngine on context + state persistence. CommunicationsEngine receives EM environment.

Gaps (documented deferrals): - CarrierOpsEngine not in tick loop (structural only) - Beaufort operational gates (helo abort, carrier suspension) need air sortie dispatch - Hull natural period hardcoded rather than from YAML

Deficits

8 new deferred items (all logged above). Key ones: - Hull natural period should come from YAML per ship class - CarrierOpsEngine battle loop integration (requires air sortie dispatch architecture) - SHF/EHF comms rain attenuation (needs per-equipment frequency tracking)

Performance

Test suite: 787.78s (13:07) — consistent with Phase 60. No performance regression.

Action Items

  • All source modifications complete
  • All tests passing (71 phase + 8,344 total)
  • Update lockstep docs (CLAUDE.md, development-phases-block7.md, devlog/index.md, README.md, MEMORY.md, project-structure.md)