Block 7: Final Engine Hardening — Integration, Wiring, & Fidelity

Motivation

Blocks 1–6 built 57 phases of subsystems, from core infrastructure through combat, C2, EW, space, CBRN, four historical eras, a web UI, and full scenario validation. The result is a simulation with ~8,655 tests, 37 validated scenarios, and zero unresolved deficits.

However, a comprehensive audit reveals that the project suffers from a systemic build-then-defer-wiring pattern. The consequence is severe:

• 36 environmental parameters are computed every tick but never consumed by any downstream system
• 16 combat engines are instantiated but unreachable from the battle loop (including ALL air combat)
• 4 EngagementType enum values (MISSILE, AIR_TO_AIR, AIR_TO_GROUND, SAM) are declared but never assigned or routed
• The event bus is observational only — 73 files publish events, zero functional subscribers react to them
• Damage detail is discarded — DamageEngine computes casualties, equipment damage, fires, and ammo cookoff, but only damage_fraction is used for a binary DESTROYED/DISABLED check
• Logistics doesn't gate combat — units with zero fuel still move, maintenance state doesn't affect readiness
• Checkpoint infrastructure saves nothing — 136 classes implement get_state/set_state but none are registered
• DUG_IN provides zero protection — only stops movement, doesn't reduce incoming damage
• Detection → AI uses ground truth — AI sees raw enemy count, not detected contacts or confidence levels

Block 7 is the final hardening block. It comprehensively addresses every integration gap — environment, combat routing, cross-module feedback loops, and dead code — with a triage framework that prevents scope creep and structural verification tests that prevent future regression.

Block 7 exit criterion: 1. Every instantiated engine either contributes to simulation outcomes or is removed 2. Every computed parameter is consumed by at least one downstream system (verified by automated test) 3. Every published event type has at least one functional subscriber (verified by automated test) 4. Every cross-module feedback loop is either fully wired or explicitly documented as deferred 5. All scenarios still produce correct historical outcomes after integration

---

Current State: Engine Wiring Audit

Fully Wired & Active (contributing to outcomes)

Engine	Lines	What It Does
WeatherEngine	289	Markov weather, wind (O-U process), visibility, temperature
TimeOfDayEngine	197	Solar/lunar illumination, thermal contrast, NVG effectiveness
SeasonsEngine	253	Ground state (frozen/thawing/wet/dry), mud, snow, vegetation, trafficability
FogOfWarManager	433	Per-side detection, track lifecycle, COP sharing
PoliticalPressureEngine	268	Escalation ladder, desperation index, war crimes cascade
HeightmapManager	260	Elevation, slope, aspect
TerrainClassification	280	Land cover classes, trafficability lookup
LOSEngine	450	DDA line-of-sight, viewshed
DetectionEngine	400+	Unified SNR detection, Kalman tracking
CombatEngines (8+)	3,000+	All domain combat resolution
MoraleEngine	600+	Continuous Markov morale, rout, rally
MovementEngine	800+	A* pathfinding, formations, domain movement

Instantiated but Output Ignored

These engines are created and updated each tick, but their computed state is never consumed by downstream systems.

Engine	Lines	Gap
SeasonsEngine	253	trafficability, mud_depth, vegetation_density, snow_depth computed but never queried by movement or detection
SeaStateEngine	218	Wave height used for gunnery dispersion only; no effect on ship movement, carrier ops, amphibious assault, mine drift
EMPropagationEngine	200+	GPS accuracy wired; radar horizon, atmospheric attenuation, EM ducting NOT used by detection or comms

Instantiated but Never Called

These engines exist on the SimulationContext, have full implementations, but no method is ever invoked from engine.py or battle.py.

Engine	Lines	What It Would Add
ObscurantsEngine	256	Smoke/dust/fog clouds with drift, dispersion, spectral blocking
ConditionsEngine	249	Facade aggregating all environment sub-engines (may be redundant)
SpaceISREngine	206	Satellite overpass detection of formations
EarlyWarningEngine	143	Ballistic missile launch detection via GEO/HEO satellites
ASATEngine	353	Anti-satellite warfare, debris cascade (Kessler syndrome)
SIGINTEngine	488	Radar/comms geolocation, traffic analysis
ECCMEngine	270	Electronic protection techniques reducing jam effectiveness
OrderPropagationEngine	315	Order delay (echelon-scaled) + misinterpretation probability
PlanningProcessEngine	564	MDMP state machine with planning phase delays
ATOPlanningEngine	459	Air tasking order generation with sortie limits
UnconventionalWarfareEngine	397	IED, guerrilla hit-and-run, human shields

Partially Wired (gate checks but no execution)

Engine	Lines	Gap
StratagemEngine	416	Eligibility evaluation works; activate_stratagem() never called
MineWarfareEngine	541	resolve_mine_encounter() called; lay_mines() never called

Dead YAML Fields

Field	Defined In	Consumed?	Notes
weight_kg	AmmoDefinition, Equipment	No	No weight-of-fire calculations
propulsion	AmmoDefinition	No	Rocket/turbojet/ramjet — no propulsion modeling
unit_cost_factor	AmmoDefinition	No	No logistics cost modeling
data_link_range	AerialUnit loader	No	UAV data link range never checked

---

Cross-Module Integration Gaps (Non-Environment)

These are systemic integration failures that affect every domain, not just environment.

Event Bus: Observational Only

73 files publish events. 2 files subscribe — the Recorder (captures all events for replay) and scenario_runner (one generic subscription). Zero modules take functional action based on event receipt.

Events published but never processed by domain logic:

Event	Publisher	Expected Subscriber	Missing Effect
CasualtyTreatedEvent	MedicalEngine	Unit strength tracker	Treated personnel never return to duty (unit stays depleted)
ReturnToDutyEvent	MedicalEngine	Unit strength tracker	Same — RTD published but nobody restores personnel
EquipmentBreakdownEvent	MaintenanceEngine	Unit readiness	Breakdown published but unit continues at full capability
MaintenanceCompletedEvent	MaintenanceEngine	Unit readiness	Repair published but readiness never restored
SupplyDeliveredEvent	SupplyEngine	Stockpile tracker	Supply arrives but stockpile may not update consumption state
RouteInterdictedEvent	DisruptionEngine	Supply routing	Route severed but no reactive rerouting or supply crisis
ConvoyDestroyedEvent	LogisticsEngine	Supply planning	Convoy loss has no feedback to logistics or C2
MoraleStateChangeEvent	MoraleEngine	AI assessment	Morale transition published but AI doesn't adapt tactics
RallyEvent	RallyEngine	Movement/engagement	Rally published but unit doesn't automatically resume operations
ATOGeneratedEvent	ATOPlanningEngine	Air ops dispatcher	ATO entries generated but never consumed by air missions
PlanningCompletedEvent	PlanningProcessEngine	OODA cycle	Planning completes but nobody reads the result
OrderIssuedEvent	OrderPropagationEngine	Unit execution	Order published but no delayed-execution queue
StratagemActivatedEvent	StratagemEngine	Combat modifier	Stratagem activates but no modifier applied

Root cause: The EventBus was designed for cross-module communication but modules were built with direct parameter passing instead. Events serve only as audit trail for the recorder.

Combat Routing: 16 Unreachable Engines

4 dead EngagementType values:

EngagementType	Status	Consequence
MISSILE	Never assigned by _infer_engagement_type()	Missile flights tracked by MissileEngine but never end in impact
AIR_TO_AIR	Never assigned	Air-to-air engagements route as DIRECT_FIRE (gun duel physics)
AIR_TO_GROUND	Never assigned	CAS/strike uses generic engagement, not air-specific Pk model
SAM	Never assigned	SAM engagements use generic engagement, not AD-specific model

16 engines instantiated but unreachable from battle loop:

Category	Engines	Lines	Impact
Air combat	AirCombatEngine, AirDefenseEngine, AirGroundEngine	~1,200	Air engagements use wrong physics (ground combat Pk for air duels)
Strategic air	AirCampaignEngine, StrategicBombingEngine, StrategicTargetingEngine	~800	Strategic air campaigns can't execute
Missiles	MissileEngine, MissileDefenseEngine	~600	Missile flight/intercept models unused
Naval air	CarrierOpsEngine	~300	CAP management, recovery windows unused
Historical naval	NavalOarEngine	~200	Ancient trireme combat uses generic model
Damage effects	IncendiaryDamageEngine, UXOEngine	~400	Fire zones and UXO contamination never created
Unconventional	UnconventionalWarfareEngine	~400	IED/guerrilla tactics dead
Siege	SiegeEngine	~300	Siege mechanics never invoked
C2	VisualSignalsEngine, AmphibiousAssaultEngine	~400	Ancient C2 and amphibious assault dead

Damage Detail Discarded

DamageEngine.resolve_damage() returns a DamageResult with: - damage_fraction (0.0–1.0) — USED: compared to destruction/disable thresholds - casualties (list of CasualtyResult) — DISCARDED: personnel losses never extracted - systems_damaged (list of equipment hits) — DISCARDED: equipment never degrades - fire_started (bool) — DISCARDED: no fire zone creation - ammo_cookoff (bool) — DISCARDED: no secondary explosions - penetrated (bool) — DISCARDED: no armor degradation

Consequence: A unit is either OPERATIONAL, DISABLED, or DESTROYED. No partial degradation. A tank with 49% damage has full combat power; at 50% it's destroyed. No crew casualties, no degraded systems, no fires.

Logistics Does Not Gate Combat

Resource	Tracked?	Consumed?	Gates Action?
Ammunition	Yes	Yes (weapon.fire())	NO — unit fires at 0 ammo
Fuel	Yes	No (movement doesn't consume)	NO — unit moves at 0 fuel
Maintenance/Readiness	Yes	Yes (breakdown events)	NO — disabled equipment still functions
Medical	Yes	Yes (treatment events)	NO — RTD events not consumed
Supply routes	Yes	Yes (disruption modeled)	NO — severed routes don't create shortages

Note: The third agent's investigation confirmed ammo consumption IS wired via weapon.fire(). However, ammo depletion does not prevent firing.

Checkpoint State Not Saved

CheckpointManager.register() exists but is never called in production code. Checkpoints contain only clock and RNG state. Module state (morale, detection tracks, supply levels, equipment condition) is not saved. This means checkpoint restore produces a simulation with correct time and RNG but wrong unit states.

Detection → AI Uses Ground Truth

The AI assessment system receives: - contacts = enemy_unit_count (raw count, not detected count) - enemy_power = float(enemies) (true strength, not estimated)

FogOfWarManager exists and tracks per-side contacts with confidence/age, but this data is not passed to the AI assessment pipeline. The AI has perfect information about enemy forces regardless of sensor coverage.

Exception: When enable_fog_of_war is True (never set by any scenario), contact count comes from FOW tracking. But even then, quality/confidence/error is not conveyed.

Posture Protection Gap

Posture	Speed Mult	Damage Reduction	Should Have
MOVING	1.0	0%	0% (correct)
HALTED	1.0	0%	0% (correct)
DEFENSIVE	0.5	0%	~20% (hasty fighting position)
DUG_IN	0.0	0%	~50% (prepared position)
FORTIFIED	0.0	0%	~70% (hardened position)

DUG_IN/FORTIFIED units take the same casualties as units in the open. The only benefit is speed=0 (they don't move). Terrain cover from obstacles provides some protection, but posture itself does not.

C2 Authority Not Enforced

• ROE: Works. Engagement gated by ROE level per target category.
• Command authority: Not enforced. Any unit engages any target regardless of command chain.
• Orders → behavior: Orders are not consumed. Units don't read orders to modify target selection or movement.
• Comms loss → independent action: Comms degradation affects assessment confidence but not engagement rules. A unit with zero comms fights identically to one with perfect comms.

CalibrationSchema Fields Never Exercised

16 calibration fields have defaults but are never set by any scenario YAML (confirmed via YAML audit):

disable_threshold, dew_disable_threshold, dig_in_ticks, wave_interval_s, target_selection_mode, night_thermal_floor, wind_accuracy_penalty_scale, rain_attenuation_factor, c2_min_effectiveness, enable_fog_of_war, engagement_concealment_threshold, target_value_weights, gas_casualty_floor, gas_protection_scaling, subsystem_weibull_shapes, victory_weights

Note: observation_decay_rate was previously listed but is set in at least one scenario YAML.

---

Triage Framework

Every identified gap must be triaged before implementation. This prevents scope creep and ensures we focus on high-impact work.

Priority Levels

Priority	Criterion	Action	Examples
P0: Must Wire	Existing code with clear consumer path; high fidelity impact; missing wiring is a bug	Wire in this block	Rain detection factor (function exists, never called), air combat routing, damage detail extraction, posture protection
P1: Should Wire	Moderate fidelity impact; straightforward integration; improves realism noticeably	Wire in this block	Seasonal trafficability → movement, obscurants → detection, radar horizon, HF quality → comms
P2: Wire If Time	Lower fidelity impact; useful but not essential; may require new code beyond simple wiring	Wire if ahead of schedule	Fire spread model, altitude sickness, wave period resonance, biological sonar noise
P3: Defer	Edge cases, very low impact, or requires significant new architecture	Document and defer to future block	Spin drift, soil CBRN absorption, dynamic cratering, Faraday rotation
P4: Remove	Computed values with no meaningful consumer; dead code that generates false audit findings	Delete computation	shadow_azimuth (no consumer), deep_channel_depth (niche), solar/lunar decomposition (diagnostic only)

Triage: Environment Parameters (36 items)

Parameter	Priority	Rationale
_compute_rain_detection_factor() — function exists, never called	P0	Bug — implemented ITU-R P.838 model, zero call sites
DEW humidity/precip_rate — params accepted, never passed	P0	Bug — physics model exists, call site doesn't pass values
SeasonsEngine.mud_depth → movement	P1	Rasputitsa, WW1 mud — major historical effect
SeasonsEngine.snow_depth → movement	P1	Eastern Front, Korean War — major historical effect
SeasonsEngine.vegetation_density → detection	P1	Seasonal concealment — moderate historical effect
SeasonsEngine.sea_ice_thickness → crossing	P1	Eastern Front frozen rivers — scenario-specific
ObscurantsEngine (full instantiation + wiring)	P1	Smoke screens, dust trails — significant tactical effect
EMPropagation.radar_horizon → detection	P1	Earth curvature on radar — fundamental physics
EMPropagation.hf_quality → comms	P1	Day/night HF reliability — affects WW2/historical comms
EMPropagation.atmospheric_attenuation → radar/comms	P1	Rain loss on high-freq systems — moderate effect
EMPropagation.ducting → radar range	P1	Maritime radar extension — scenario-specific
UnderwaterAcoustics.thermocline → sonar	P1	ASW layer tactics — fundamental naval effect
UnderwaterAcoustics.convergence_zones → sonar	P1	55km detection rings — fundamental ASW
SeaState.tidal_current → ship movement	P1	Ship routing, mine drift — moderate effect
TimeOfDay.background_temp → thermal ΔT detection	P1	Thermal crossover vulnerability — real planning factor
TimeOfDay.nvg_effectiveness → detection	P1	NVG-equipped night detection — moderate effect
WeatherEngine.wind.gust → operation gates	P2	Helicopter abort, parachute accuracy — niche
WeatherEngine.pressure → air density	P2	Ideal gas correction — small effect at sea level
SeasonsEngine.vegetation_moisture → fire	P2	Fire ignition probability — requires fire system
SeasonsEngine.wildfire_risk → fire events	P2	Spontaneous fire — requires fire system
SeasonsEngine.daylight_hours → ops planning	P2	Operational tempo — low tactical impact
SeaState.wave_period → ship resonance	P2	Roll amplitude — niche effect
SeaState.beaufort → small craft gate	P2	Landing craft operations — scenario-specific
UnderwaterAcoustics.surface_duct → sonar	P2	In-duct detection — niche ASW
Terrain.vegetation_height → LOS	P2	Ground-level vegetation block — moderate
Terrain.combustibility → fire	P2	Fire ignition — requires fire system
Obstacles.traversal_risk → casualties	P2	Wire crossing casualties — niche
Obstacles.traversal_time → movement	P2	Obstacle delay — moderate
Hydrography.ford_points → crossing	P2	River ford routing — scenario-specific
Infrastructure.bridge_capacity → weight gate	P2	Heavy vehicle bridge limit — niche
TimeOfDay.shadow_azimuth	P4	No meaningful consumer — remove
TimeOfDay.solar/lunar contribution split	P4	Diagnostic only — remove computation
UnderwaterAcoustics.deep_channel_depth	P3	SOFAR channel — very niche
Equipment.temperature_range → stress	P2	Weapon jam in extreme cold/heat — moderate
Infrastructure.Road.speed_factor override	P3	Per-road speed — low impact vs hardcoded table
Infrastructure.Tunnel routing	P3	Tunnel in pathfinding — niche

Triage: Combat Integration (16+ items)

Gap	Priority	Rationale
Air combat routing (AIR_TO_AIR, AIR_TO_GROUND, SAM)	P0	3 engines exist with correct physics; currently air duels use ground combat Pk
Damage detail extraction (casualties, systems_damaged)	P0	DamageEngine computes it; discarding is a bug
Posture → damage reduction (DUG_IN/FORTIFIED protection)	P0	DUG_IN providing zero protection is a bug
MISSILE engagement routing	P1	MissileEngine exists; missiles currently route as DIRECT_FIRE or COASTAL_DEFENSE
IncendiaryDamageEngine → fire zones	P1	Engine exists; fire_started from DamageResult should trigger it
CarrierOpsEngine → naval air	P1	CAP management, recovery windows — significant for naval scenarios
SiegeEngine → campaign loop	P2	Ancient/medieval siege — era-specific
AmphibiousAssaultEngine → naval ops	P2	Beach assault state machine — scenario-specific
UXOEngine → post-engagement	P2	Submunition contamination — niche
NavalOarEngine → ancient naval	P3	Ancient trireme propulsion — era-specific niche
VisualSignalsEngine → ancient C2	P3	Ancient visual communication — era-specific niche
StrategicBombingEngine → WW2 campaign	P2	Strategic air campaign — era-specific
StrategicTargetingEngine → bombing	P2	Target prioritization — paired with StrategicBombing
AirCampaignEngine → campaign loop	P2	Campaign-level air operations — moderate
MissileDefenseEngine → AD	P2	Air defense missiles — moderate

Triage: Cross-Module Integration

Gap	Priority	Rationale
Fuel consumption → movement gate	P0	Fundamental logistics — units at 0 fuel should not move
Ammo depletion → firing gate	P0	Unit fires at 0 ammo — bug
Maintenance/readiness → combat gate	P1	Disabled equipment should degrade combat performance
Medical RTD → unit strength restoration	P1	Treated casualties should return to duty
Detection → AI assessment (use FOW contacts, not ground truth)	P1	AI has perfect information — defeats purpose of detection system
Comms loss → C2 degradation → independent action	P1	Comms down should force units to last received orders
Checkpoint state registration	P1	Checkpoint saves nothing — defeats purpose of checkpoint system
Event bus functional subscribers	P2	Most events are consumed via direct passing; event subscribers are optional improvement
Supply route interdiction → supply shortage	P2	Logistics feedback loop — moderate
CalibrationSchema exercised in scenarios	P2	16 fields never set — either exercise or remove

---

Structural Verification Tests

These tests must be created before implementation begins and run continuously to prevent regression.

Test 1: Unconsumed Parameter Audit

def test_no_unconsumed_engine_outputs():
    """Every engine property computed during simulation is read by at least one consumer."""
    # Instrument all engine classes with property access tracking
    # Run a representative scenario with all systems active
    # Collect set of properties that were written but never read
    # Assert empty set (or only P4-triaged items)

Implementation: Monkey-patch engine classes with __getattr__/__setattr__ tracking, or use a coverage-like instrumentation approach. Run a modern scenario + one historical scenario to exercise all code paths.

Test 2: Dead Method Audit

def test_no_uncalled_public_methods():
    """Every public method on every engine class has at least one external caller."""
    # AST-parse all source files
    # Build method definition set (all public methods on engine classes)
    # Build call site set (all method calls across all source files)
    # Assert: definitions ⊆ call sites (minus test-only callers)

Implementation: Use ast.parse() to walk source trees. Flag any public method defined in stochastic_warfare/ that has zero call sites outside its own module and tests.

Test 3: Event Subscription Audit

def test_all_event_types_have_subscribers():
    """Every event type published in the simulation has at least one functional subscriber."""
    # AST-parse all source files for bus.publish(EventType(...))
    # AST-parse all source files for bus.subscribe(EventType, handler)
    # Assert: every published type has at least one subscriber (beyond Recorder)
    # OR: published type is in OBSERVATION_ONLY_EVENTS allowlist

Test 4: Engagement Routing Completeness

def test_all_engagement_types_routed():
    """Every EngagementType enum value has a handler in the battle loop."""
    # For each EngagementType value
    # Assert: route_engagement() or _infer_engagement_type() can produce it
    # AND: a handler exists that resolves it

Test 5: Feedback Loop Verification

def test_logistics_gates_combat():
    """Unit with depleted ammo cannot fire; unit with depleted fuel cannot move."""

def test_damage_detail_applied():
    """DamageResult casualties are extracted and reduce unit personnel count."""

def test_posture_protection():
    """DUG_IN unit takes less damage than MOVING unit from same attack."""

def test_checkpoint_state_round_trip():
    """Checkpoint saves and restores all module state, not just clock/RNG."""

---

Theme 1: Environmental Fidelity Gap Analysis (Priority: Highest)

The environment subsystem has rich physics models whose outputs are largely ignored. This section is an exhaustive, domain-by-domain audit of every environmental parameter — what's computed, what's consumed, and what's missing entirely. The goal is "excruciatingly exact" fidelity: every computed value consumed, every real-world effect that matters at campaign/battle scale represented.

Legend

• WIRED: Computed and consumed by downstream combat/movement/detection systems
• COMPUTED/UNCONSUMED: Engine calculates the value every tick, but no system reads it
• DEFINED/UNCALLED: Method or API exists in source code but is never invoked
• NOT IMPLEMENTED: Physics model doesn't exist yet but should for high fidelity

---

1.1 Atmospheric Physics

Air Density & Pressure

Parameter	Status	Current State	What Should Happen
Air density (altitude-dependent)	WIRED (partial)	Exponential model in ballistics (scale height 8500m). Used for Mach drag.	Also needed for: aircraft engine performance, helicopter lift margin, DEW thermal blooming, unguided rocket range
Barometric pressure	NOT IMPLEMENTED	WeatherEngine computes pressure field but it's never consumed anywhere	Should feed air density via ideal gas law (ρ = P/RT). Affects ballistic drag, acoustic propagation, altimeter accuracy
Humidity → air density	NOT IMPLEMENTED	Humidity computed in WeatherEngine, never affects air density	Moist air is ~0.5% less dense than dry air. Small effect but contributes to density altitude calculation
Temperature lapse rate	NOT IMPLEMENTED	Ballistics uses constant temperature at all altitudes	ISA lapse rate (−6.5°C/km to tropopause, isothermal above). Affects speed of sound, Mach number, drag coefficient at altitude
Density altitude	COMPUTED/UNCONSUMED	ConditionsEngine.air().density_altitude computed	Should gate: helicopter operations (high+hot = no-go), aircraft takeoff distance, engine power output

Wind

Parameter	Status	Current State	What Should Happen
Mean wind (speed, direction)	WIRED	Ornstein-Uhlenbeck process, consumed by crosswind penalty, CBRN drift, obscurant drift	—
Wind gusts	COMPUTED/UNCONSUMED	WeatherEngine.wind.gust computed every tick	Should affect: helicopter operations (gust > threshold = abort landing), bridge-laying operations, parachute drop accuracy, amphibious small craft capsizing
Wind shear (altitude-dependent)	NOT IMPLEMENTED	Wind is constant at all altitudes	Real wind varies with altitude (boundary layer). Affects artillery at high angles, parachute drift, aircraft approach. Model: log wind profile or power law
High wind halt threshold	NOT IMPLEMENTED	No maximum wind speed for operations	Should force infantry to halt in extreme wind (>25 m/s / ~50 mph), prevent helicopter operations, degrade vehicle stability
Turbulence	NOT IMPLEMENTED	Only mean wind modeled	Adds stochastic dispersion to ballistic trajectories, parachute drops, helicopter hover stability. Model: σ_wind as fraction of mean wind

Temperature

Parameter	Status	Current State	What Should Happen
Ambient temperature	WIRED	Diurnal cycle + monthly mean. Fed to ballistics (speed of sound), weather state transitions	—
Temperature at altitude	COMPUTED/UNCONSUMED	WeatherEngine.temperature_at_altitude() exists	Should drive lapse rate calculations, icing determination, engine performance curves
Temperature inversion	NOT IMPLEMENTED	No inversion layer detection	Critical for: CBRN (traps agents below inversion, 10x concentration increase), EM ducting (extends radar range), acoustic propagation (sound channels). Model: detect when temperature increases with altitude
Propellant temperature → muzzle velocity	NOT IMPLEMENTED	All rounds assume nominal MV	Cold propellant (−20°C) reduces MV by 2–5%. Hot propellant (+50°C) increases MV by 1–3%. Affects range tables for artillery. Source: MIL-STD-1474
Equipment temperature stress	DEFINED/UNCALLED	EquipmentItem.temperature_range defined in YAML, EquipmentManager.environment_stress() computes degradation factor	Degradation factor never applied in simulation loop. Should affect: weapon reliability (jams in extreme cold/heat), electronics failure rate, battery capacity

Icing

Parameter	Status	Current State	What Should Happen
Icing risk	COMPUTED/UNCONSUMED	ConditionsEngine.air().icing_risk calculated (temperature 0 to −20°C + cloud/precip)	Should affect: aircraft performance (wing ice = lift loss, engine ice = power loss), helicopter blade ice, radar dome ice (degrades signal), weapon system freeze-up
Road/runway icing	NOT IMPLEMENTED	No surface ice model	Snow/ice on runways should degrade aircraft operations. Road ice should reduce vehicle traction. Model: surface temperature < 0°C + precipitation = ice

---

1.2 Optical & Visual Environment

Illumination & Visibility

Parameter	Status	Current State	What Should Happen
Total illumination (lux)	WIRED	Solar + lunar + cloud cover → detection range modifier	—
Twilight stages (5-level)	WIRED	Civil/nautical/astronomical/full night → graduated visual/thermal modifiers	—
Weather visibility	WIRED	Weather state → visibility_m → detection range cap	—
Solar contribution (decomposed)	COMPUTED/UNCONSUMED	Separated from total lux but only total used	Low value — decomposition is diagnostic data
Lunar contribution (decomposed)	COMPUTED/UNCONSUMED	Same as solar	Low value
Artificial illumination (flares, searchlights)	DEFINED/UNCALLED	API exists in TimeOfDayEngine (artificial_contribution) but never populated	Should allow: flare illumination events (brief high-lux area), searchlight beams, urban ambient light. Model: point source with inverse-square falloff
Shadow azimuth	COMPUTED/UNCONSUMED	Computed from solar position, never queried	Could affect: visual detection from specific angles (target in shadow harder to spot from shadow side). Low priority
NVG effectiveness	COMPUTED/UNCONSUMED	TimeOfDayEngine.nvg_effectiveness() exists	Partially consumed by movement (0.7x night speed recovery with NVG). NOT consumed by detection engine — NVG-equipped units should detect further at night

Thermal Environment

Parameter	Status	Current State	What Should Happen
Thermal contrast	WIRED (partial)	TimeOfDayEngine.thermal_environment().thermal_contrast computed. Detection uses illumination-based modifier instead	Should be the primary driver of thermal detection SNR. Vehicles have high thermal contrast at night (engine heat vs cold background), low at dawn/dusk crossover
Background temperature	COMPUTED/UNCONSUMED	Computed as ambient + solar heating + surface material	Should feed thermal detection SNR calculation: target_temp − background_temp = ΔT → SNR. Currently uses generic thermal_contrast scalar instead of physics-based ΔT
Thermal crossover timing	COMPUTED/UNCONSUMED	crossover_in_hours computed — tells how long until thermal signatures blend into background	Should create vulnerability windows: at crossover, thermal detection is near-zero for stationary vehicles. Dawn and dusk crossovers are real-world tactical planning factors

Obscurants (Smoke, Dust, Fog)

Parameter	Status	Current State	What Should Happen
Smoke cloud deployment	DEFINED/UNCALLED	Full API: deploy_smoke(center, radius, multispectral) in ObscurantsEngine (256 lines)	Barrages/mortars should spawn smoke at impact. Smoke grenades/pots for deliberate screening. Spectral blocking: visual 0.9, thermal 0.1 (standard) or 0.8 (multispectral), radar 0.0–0.3
Smoke drift & decay	DEFINED/UNCALLED	Wind-driven drift, radial dispersion (r ∝ √t), exponential decay (30min half-life) all implemented	ObscurantsEngine.update() should be called each tick. Currently never instantiated on SimulationContext
Dust from vehicle movement	NOT IMPLEMENTED	No dust generation hook in movement	Vehicles on dry terrain (DRY ground_state + non-road) should spawn dust trails. Intensity ∝ speed × vehicle_count. Dust reveals movement (signature enhancement) + degrades following-vehicle LOS
Dust from explosions	NOT IMPLEMENTED	Impact detonations don't create dust	HE impacts on dry ground should create short-duration dust clouds at impact site. Affects BDA (can't see if target was hit)
Fog auto-generation	DEFINED/UNCALLED	ObscurantsEngine has fog generation when WeatherState.FOG	Never triggered because engine never instantiated. Should auto-create fog patches when weather transitions to FOG. Radiation fog (clear nights), advection fog (warm air over cold water), sea fog
Fog dissipation by solar heating	NOT IMPLEMENTED	Fog uses generic exponential decay	Real fog lifts when solar heating warms ground (morning burn-off). Model: fog opacity × (1 − solar_heating_factor)
Spectral opacity query	DEFINED/UNCALLED	opacity_at(pos) returns per-spectrum opacity (visual, thermal, radar)	Should be queried by: detection engine (degrade SNR by spectrum), combat engine (reduce Pk for visual/thermal-guided weapons), DEW engine (attenuate beam)
Pre-placed smoke/fog in scenarios	NOT IMPLEMENTED	No scenario YAML field for initial obscurant placement	Allow environment_config.smoke_zones and fog_zones for historical accuracy (Austerlitz morning fog, Falklands sea fog, WW1 smoke barrages)

---

1.3 Electromagnetic Environment

Parameter	Status	Current State	What Should Happen
GPS accuracy	WIRED	EMPropagationEngine → detection/space/EW integration	—
GPS spoofing offset	WIRED	EW module → EMPropagation → position error	—
GPS jam degradation	WIRED	EW module → EMPropagation → accuracy degradation	—
Free-space path loss	COMPUTED/UNCONSUMED	free_space_path_loss(freq, range) public API exists, only called internally	Should be used by: communications range calculation, radar detection range, EW intercept range
Radar horizon	DEFINED/UNCALLED	radar_horizon(antenna_height) computes 4/3 Earth radius geometric horizon	Should gate air defense radar detection range. Currently detection uses fixed max_range_m from sensor YAML, ignoring Earth curvature. A ground radar at 10m height has ~13km horizon; at 30m, ~22km. Low-flying aircraft below horizon are invisible to radar
Atmospheric attenuation (freq-dependent)	COMPUTED/UNCONSUMED	Computed per frequency band (O₂ absorption peak at 60 GHz, water vapor at 22 GHz)	Should attenuate radar and comms at high frequencies. X-band (10 GHz) loses ~0.01 dB/km dry, ~0.1 dB/km in rain. Ka-band (35 GHz) loses ~0.4 dB/km in rain
EM ducting (super-refraction)	COMPUTED/UNCONSUMED	ducting_possible and refraction factor computed based on humidity + temperature gradient	Should extend radar and comms range beyond geometric horizon in warm, humid maritime environments (common in Persian Gulf, South China Sea). Factor of 2–3x range extension
Duct height	COMPUTED/UNCONSUMED	Computed but never read	Determines which systems benefit from ducting (antenna must be within duct layer)
HF propagation quality (day/night)	COMPUTED/UNCONSUMED	hf_propagation_quality() computes D-layer absorption (day) vs F-layer reflection (night)	Should modulate HF radio reliability in communications engine. Day: D-layer absorbs → short HF range. Night: F-layer reflects → long HF range (skip propagation). Critical for WW2, some modern scenarios
Rain clutter on radar	DEFINED/UNCALLED	_compute_rain_detection_factor(precip_rate, range) implemented in battle.py with ITU-R P.838 model but never called	Should reduce radar detection probability in rain. Heavy rain (25 mm/hr) creates radar returns that mask targets. Already implemented — just needs a call site
Frequency-dependent rain attenuation for comms	NOT IMPLEMENTED	Generic comms model doesn't consider frequency	UHF (300 MHz–3 GHz): negligible rain attenuation. SHF (3–30 GHz): significant. EHF (30–300 GHz): severe. Should vary comms reliability by frequency band in rain
Tropospheric ducting for radio	NOT IMPLEMENTED	Only modeled for radar, not comms	Same physics extends radio comms range in certain weather conditions. Common in maritime operations
Ionospheric storm effects	NOT IMPLEMENTED	No space weather model	Solar flares and geomagnetic storms degrade HF comms for hours to days. Model: random events with probability ∝ solar cycle
Radio horizon (altitude-dependent)	NOT IMPLEMENTED	Only radar_horizon() exists for radar	Identical physics applies to radio — higher antenna = longer range. Mobile units at different altitudes have different comms range. Model: same as radar_horizon but for comms

---

1.4 Acoustic Environment

Parameter	Status	Current State	What Should Happen
Sound velocity profile (SVP)	WIRED	Mackenzie equation (T, S, depth) → sonar detection	—
Ambient noise (Wenz curves)	WIRED	Beaufort → dB ambient noise → sonar SNR threshold	—
Transmission loss	WIRED	Spherical spreading + absorption → sonar range	—
Surface duct depth	COMPUTED/UNCONSUMED	Calculated from temperature profile but never used	Sonar trapped in surface duct has enhanced range within duct but cannot detect below it. Submarine can hide below duct. Model: if target_depth > duct_depth, add 20 dB loss
Thermocline depth	COMPUTED/UNCONSUMED	Calculated but never used in detection	Layer below which sound bends downward. Submarine below thermocline is very difficult to detect from surface sonar. Critical for ASW tactics
Deep sound channel depth	COMPUTED/UNCONSUMED	Calculated but never used	SOFAR channel enables detection at extreme ranges (hundreds of km). Low priority — mostly for fixed sonobuoy networks
Convergence zone ranges	COMPUTED/UNCONSUMED	convergence_zone_ranges(source_depth) calculates 55km interval CZ detections	Should create detection "rings" at 55km, 110km, 165km from sonar source. Target between zones is in shadow. Fundamental to deep-water ASW
Shipping noise	NOT IMPLEMENTED	Only wind-driven Wenz curve ambient noise	Busy shipping lanes increase ambient noise, degrading sonar. Model: proximity to shipping lanes → +5–15 dB ambient noise
Biological noise	NOT IMPLEMENTED	No marine life noise model	Whale song, snapping shrimp can mask sonar contacts in certain waters/seasons. Low priority — niche effect

---

1.5 Ground & Terrain Environment

Seasonal Ground State

Parameter	Status	Current State	What Should Happen
Ground trafficability (seasonal)	WIRED (partial)	SeasonsEngine → ground_trafficability scalar consumed by MovementEngine	Currently uses coarse lookup: FROZEN=0.9, THAWING=0.3, WET=0.6, DRY=1.0, SATURATED=0.2, SNOW=0.5. Wired but should also differentiate wheeled vs tracked (wheeled suffer more in mud)
Mud depth	COMPUTED/UNCONSUMED	SeasonalConditions.mud_depth accumulated from rain + thaw	Should provide granular speed penalty beyond binary trafficability. Deep mud (>20cm) immobilizes wheeled vehicles entirely. Tracked vehicles at 50% speed. Logistics transport has mud_speed_fraction=0.5 but battle movement ignores mud depth
Snow depth	COMPUTED/UNCONSUMED	SeasonalConditions.snow_depth accumulated from snowfall	Deep snow (>30cm) should slow infantry to 40% speed, vehicles to 60%. Snow on roads should degrade road speed bonus. Plowed roads partially recover. Ski troops should have reduced penalty
Vegetation density	COMPUTED/UNCONSUMED	Sigmoid of growing-degree-days, 0.0–1.0	Should modify: visual detection concealment (summer foliage hides, winter bare exposes), movement speed through dense vegetation, fire spread rate
Vegetation moisture	COMPUTED/UNCONSUMED	Tracked alongside vegetation density	Should affect: fire spread probability (dry vegetation burns easily, wet doesn't), agent persistence (moist surfaces break down chemical agents faster)
Sea ice thickness	COMPUTED/UNCONSUMED	Accumulated from freezing degree-days, latitude > 50°	Should enable: ice crossing for infantry (>10cm), light vehicles (>30cm), heavy vehicles (>60cm). Model: weight-dependent ice failure threshold. Critical for Eastern Front (crossing frozen rivers), Finnish Winter War, Korean War Chosin Reservoir
Wildfire risk	COMPUTED/UNCONSUMED	Composite of dryness × heat × wind × low humidity	Should trigger: random wildfire events in high-risk areas, fire spread after incendiary strikes. Feeds fire zone system
Daylight hours	COMPUTED/UNCONSUMED	From astronomy day_length_hours	Should affect: operational planning (short winter days = less time for maneuver), fatigue accumulation rate, solar panel power for unmanned systems
Rasputitsa (spring mud season)	PARTIALLY IMPLEMENTED	Mud depth mechanics exist but transitions not dramatic enough	Historical spring mud seasons halted entire armies (Eastern Front 1941, 1942, 1943, 1944). Thawing ground + rain should create impassable conditions lasting weeks. Model: THAWING state + rain → SATURATED with mud_depth spike

Terrain Properties

Parameter	Status	Current State	What Should Happen
Vegetation height	COMPUTED/UNCONSUMED	TerrainProperties.vegetation_height per land cover class	Should affect: visual detection (tall vegetation conceals vehicles at distance), LOS check (thick vegetation blocks LOS at ground level), helicopter landing clearance
Combustibility	COMPUTED/UNCONSUMED	TerrainProperties.combustibility per land cover class (e.g., dry_grass=0.9, urban=0.2)	Should feed fire zone creation: incendiary weapons on high-combustibility terrain create fires; on low-combustibility terrain they don't. Currently defined but never read
Obstacle traversal risk	DEFINED/UNCALLED	Obstacle.traversal_risk defined in obstacle model	Should apply: casualty probability when crossing obstacles (e.g., wire, minefield, rubble). Currently obstacles are binary pass/block
Obstacle traversal time	DEFINED/UNCALLED	Obstacle.traversal_time_multiplier defined	Should slow units crossing obstacles (wire obstacles = 3x time, rubble = 2x, water = 4x). Currently hardcoded in movement
Ford points	DEFINED/UNCALLED	Hydrography.is_fordable(), ford_points_near() implemented	Should enable river crossing at ford points with depth/current-dependent time cost. Currently river crossing is binary passable/not-passable
Bridge capacity	DEFINED/UNCALLED	Bridge.capacity_tons defined	Should gate heavy vehicles on bridges (MBT at 60+ tons can't cross a 20-ton bridge). Currently all bridges passable by all units
Tunnel routing	DEFINED/UNCALLED	Tunnel geometry defined in infrastructure	Should be available in pathfinding for protection from air attack. Currently never routed
Road speed factor override	DEFINED/UNCALLED	Road.speed_factor field exists	Currently uses hardcoded _ROAD_SPEED_FACTORS dict instead of per-road values. Should use road-specific factors (highway vs dirt road vs track)
Soil type → dig-in time	NOT IMPLEMENTED	All terrain has same dig-in rate	Sandy soil: fast digging, poor fortification. Rocky soil: slow digging, good fortification. Frozen ground: very slow digging. Clay: moderate. Model: soil_type multiplier on dig_in_ticks
Dynamic terrain (cratering)	NOT IMPLEMENTED	Heightmap is static	Heavy artillery should create craters that affect movement and provide cover. Low priority — high implementation cost (pathfinding cache invalidation)

---

1.6 Maritime Environment

Parameter	Status	Current State	What Should Happen
Wave height (Pierson-Moskowitz)	WIRED	Sig. wave height → naval gunnery dispersion, carrier bolter probability, amphibious penalty	—
Wave period	COMPUTED/UNCONSUMED	Computed from wave spectrum	Should affect: ship resonance (period matching hull natural frequency = extreme roll), helicopter deck landing difficulty. Model: roll amplitude ∝ 1/(
Tide height	WIRED (partial)	Harmonic model (M2/S2/K1/O1), consumed by amphibious calculations	Should also affect: shallow-water navigation (ground contact risk), beach approach timing, submarine depth clearance in littoral waters
Tidal current (speed, direction)	COMPUTED/UNCONSUMED	Computed from tidal harmonics but never consumed	Should affect: ship movement speed (with/against current), mine drift (floating mines move with current), submarine station-keeping effort, amphibious assault beach approach vector
SST (sea surface temperature)	WIRED (partial)	Fed to underwater acoustics SVP calculation	—
Beaufort scale	WIRED (partial)	Drives sonar ambient noise (Wenz curves)	Should also affect: small craft operations (Beaufort > 5 = dangerous for landing craft), deck operations, helicopter landing, radar sea clutter
Sea spray/salt fog	NOT IMPLEMENTED	No maritime atmospheric effect	Sea spray in high winds creates: visual obscuration (like fog, reduces visibility to 1–3 km), DEW attenuation (10× worse than clean air — salt crystals scatter laser), radar clutter (close-range noise from spray droplets), corrosion (long-term equipment degradation)
Swell direction vs ship heading	NOT IMPLEMENTED	Wave height is scalar, no direction	Beam seas (waves perpendicular to ship) cause maximum roll; head seas cause pitching; following seas are smoothest. Affects gunnery accuracy, helicopter ops, crew fatigue. Model: roll_factor = sin²(wave_dir − ship_heading)
Polar ice navigation	NOT IMPLEMENTED	No ice obstacle for ships	Ice fields should block or slow ships without icebreaker capability. Ice thickness from SeasonsEngine.sea_ice_thickness. Only relevant for Arctic/Antarctic scenarios
Littoral current effects	NOT IMPLEMENTED	No coastal current model	Near-shore currents affect: mine drift, amphibious approach, submarine navigation. Could use tidal current as proxy in shallow water

---

1.7 CBRN Environment Interaction

Parameter	Status	Current State	What Should Happen
Wind advection of puffs	WIRED	Gaussian puff center moves with wind vector	—
Pasquill-Gifford stability class	WIRED	Cloud cover + wind speed + time of day → stability (A–F)	—
Terrain channeling	WIRED	Valley concentration (+50%), ridge deflection (×0.5)	—
Rain washout of agents	NOT IMPLEMENTED	Rain has no effect on airborne CBRN concentration	Rain scavenges particles and droplets from air. Model: concentration × exp(−washout_coeff × rain_rate × dt). Washout coefficient ~10⁻⁴ per mm/hr for particles. 30 mm/hr rain removes ~30% of agent in 30 minutes
Temperature → agent persistence	NOT IMPLEMENTED	Agent decay rate is constant	Nerve agents: half-life 8 hrs at 20°C, 2 hrs at 40°C, 24+ hrs at 0°C. Mustard: persists for days/weeks in cold. Model: decay_rate × exp(−Ea/RT) Arrhenius kinetics
UV hydrolysis (sunlight degradation)	NOT IMPLEMENTED	No solar degradation of agents	Direct sunlight breaks down many chemical agents. Agent concentration should decrease faster in daytime CLEAR weather. Model: +solar_degradation when is_day and cloud_cover < 0.5
Temperature inversion trapping	NOT IMPLEMENTED	No inversion detection	Agent trapped below temperature inversion layer reaches 5–10× higher concentration than open-air dispersal. Critical for valley/urban releases at night. Model: if inversion detected, multiply concentration by trapping_factor
Surface roughness → mixing height	NOT IMPLEMENTED	Single mixing height for all terrain	Urban terrain has higher surface roughness → lower effective stack height → higher ground-level concentration. Open terrain has lower roughness → better vertical mixing → lower concentration. Model: roughness_length per terrain type
Soil absorption of agents	NOT IMPLEMENTED	No ground interaction	Liquid agents pool on ground, creating contact hazard distinct from inhalation. Absorption rate depends on soil porosity. Model: deposition_velocity × concentration → ground_contamination_flux
Agent re-aerosolization	NOT IMPLEMENTED	Once deposited, agent is gone	Wind and vehicle traffic can re-aerosolize deposited agents from ground surfaces. Creates secondary hazard hours/days after initial release. Model: re-emission_rate ∝ wind_speed × ground_contamination

---

1.8 Equipment & Materiel Environment Interaction

Parameter	Status	Current State	What Should Happen
Equipment temperature range	DEFINED/UNCALLED	EquipmentItem.temperature_range = (-40, 50) defined, environment_stress() computes degradation	Degradation factor never applied. Should increase weapon jam rate, electronics failure rate, engine stall probability when temperature outside rated range
Propellant temperature → MV	NOT IMPLEMENTED	Constant muzzle velocity	Cold propellant (−20°C): −2 to −5% MV. Hot propellant (+50°C): +1 to +3% MV. Directly affects range. Source: MIL-STD-1474. Model: MV × (1 + temp_coefficient × ΔT) where temp_coefficient ≈ 0.001/°C
Lubricant viscosity	NOT IMPLEMENTED	Constant weapon reliability	Cold thickens lubricant → increased malfunction rate. Desert heat thins lubricant → increased wear. Model: reliability_modifier = f(temperature, lubricant_type)
Battery capacity at temperature	NOT IMPLEMENTED	Constant electronics availability	Li-ion at −20°C: ~50% capacity. Affects: radio endurance, NVG duration, GPS receiver, UAV flight time. Model: capacity × battery_temp_curve(T)
Fuel viscosity/gelling	NOT IMPLEMENTED	Constant fuel availability	Diesel gels at −10 to −20°C without arctic additives. Jet fuel (JP-8) freeze point −47°C. Model: fuel_available = temperature > gel_point
Engine performance vs altitude	COMPUTED/UNCONSUMED	ConditionsEngine.air().density_altitude exists	Naturally aspirated engines lose ~3% power per 300m altitude. Turbocharged engines less affected. Turbofan thrust decreases linearly above tropopause. Model: power × (1 − altitude_derating × alt/ref_alt)
Dust ingestion on engines	NOT IMPLEMENTED	No desert reliability penalty	Sand/dust degrades engine filters → power loss → eventual failure. Desert operations have 3–5× higher maintenance rates. Model: reliability_modifier in dusty terrain
Corrosion in maritime environment	NOT IMPLEMENTED	No salt spray effect	Salt air increases equipment failure rate. Extended maritime operations degrade weapons/electronics. Very low priority — long time scale

---

1.9 Human Factors & Environment

Parameter	Status	Current State	What Should Happen
Heat stress / heat casualties	NOT IMPLEMENTED	Temperature has no effect on personnel	WBGT > 32°C (wet-bulb globe temperature) + physical exertion → heat casualties. Model: heat_casualty_rate ∝ f(temperature, humidity, MOPP_level, exertion). MOPP-4 in hot weather can cause 10–20% non-combat casualties per hour
Cold injury (frostbite/hypothermia)	NOT IMPLEMENTED	Temperature has no effect on personnel	Wind chill < −25°C with prolonged exposure → cold casualties. Model: cold_casualty_rate ∝ f(wind_chill, exposure_time, clothing_level). Critical for: Chosin Reservoir, Stalingrad, Eastern Front winter
Altitude sickness	NOT IMPLEMENTED	No altitude effect on personnel performance	Above 2500m: performance degrades. Above 4000m: incapacitation without acclimatization. Model: performance_modifier = max(0.5, 1 − 0.03 × (altitude − 2500)/100) above 2500m
Fatigue from environmental stress	PARTIALLY WIRED	Movement fatigue modeled (speed reduction), not temperature/altitude driven	Temperature extremes and altitude should accelerate fatigue. Model: fatigue_rate × environmental_stress_factor where stress = f(temp_deviation_from_comfort, altitude, MOPP_level)
MOPP degradation	PARTIALLY WIRED	MOPP reduces movement speed	Should also reduce: visual detection (hood restricts FOV), manual dexterity (gloves reduce reload speed), communications (mask muffles voice), thermal stress (accelerates heat casualties). Currently only speed reduction
Wind chill	NOT IMPLEMENTED	No wind chill calculation	Wind chill = 13.12 + 0.6215T − 11.37V^0.16 + 0.3965TV^0.16. Drives cold injury rate. Model: straightforward formula from T and wind_speed
Dehydration in hot environments	NOT IMPLEMENTED	No water consumption model	Hot + dry + high exertion = 1L/hr water consumption. Running out degrades performance → casualties. Mostly a logistics concern, but affects operational tempo

---

1.10 Ballistics Environment Interaction

Parameter	Status	Current State	What Should Happen
Wind deflection on projectile	WIRED	RK4 integrator includes wind velocity vector	—
Mach-dependent drag	WIRED	Temperature-dependent speed of sound → Mach → drag multiplier	—
Air density at altitude	WIRED (partial)	Exponential atmosphere model used in drag calculation	Should also use pressure and humidity for true density (ideal gas + humidity correction). Currently uses fixed scale-height model
Coriolis effect	WIRED (optional)	Earth rotation deflection, gated by flag	—
Humidity → drag	NOT IMPLEMENTED	Constant air composition assumed	~0.5% effect on density; negligible for most weapons, measurable for long-range artillery. Low priority
Spin drift (gyroscopic precession)	NOT IMPLEMENTED	No spin-stabilized projectile drift	Rifled projectiles drift ~0.1 mil per 1000m range. Accumulates over long range. Model: drift_angle = spin_drift_constant × range. Low priority — small compared to wind
Rain impact on trajectory	NOT IMPLEMENTED	Rain doesn't affect ballistics	Heavy rain creates slight deceleration on slow projectiles (mortar). Negligible for high-velocity rounds. Very low priority
DEW thermal blooming	NOT IMPLEMENTED	No self-defocus at high power	High-power laser heats air along beam path → refractive index change → beam spreads. Model: blooming_factor = (1 + power/P_critical × range²/range_ref²). Limits effective DEW range in warm/humid air
DEW sea spray attenuation	NOT IMPLEMENTED	No maritime-specific DEW degradation	Sea spray extinction ~10 dB/km vs 0.2 dB/km in clean air. Makes shipboard laser weapons much less effective in high sea states. Model: maritime_extinction = base_extinction + sea_spray_factor(beaufort)
DEW dust attenuation	NOT IMPLEMENTED	Dust treated same as rain	Dust is ~10× more attenuating than rain at same precipitation-equivalent. Desert dust storms should severely degrade DEW. Model: dust_extinction = dust_density × mass_extinction_coefficient
Rain/fog/dust → DEW atmospheric transmittance	PARTIALLY WIRED	DEW engine accepts humidity/precip_rate parameters	Battle loop never passes environmental parameters to DEW engine. The physics model exists — the call site doesn't pass the values

---

1.11 Communications Environment Interaction

Parameter	Status	Current State	What Should Happen
Terrain LOS for comms	WIRED	Diffraction loss (6 dB) when terrain blocks LOS	—
Distance path loss	WIRED	Range-dependent link budget	—
HF propagation quality	COMPUTED/UNCONSUMED	EMPropagation.hf_propagation_quality() computes D-layer/F-layer effects by time of day	Should modulate HF radio reliability. Day: D-layer absorbs → short range only. Night: F-layer reflects → long range (skip propagation). Transforms HF comms from constant-reliability to diurnally-variable
Frequency-dependent rain attenuation	NOT IMPLEMENTED	Generic comms model ignores frequency	UHF (300 MHz – 3 GHz): negligible rain loss. SHF (3–30 GHz): 0.1–10 dB/km. EHF (30–300 GHz): severe. SATCOM uplinks (SHF/EHF) should degrade in rain. Model: ITU-R P.838 attenuation per frequency band
Radio horizon (altitude-dependent)	NOT IMPLEMENTED	Comms range is fixed	Same 4/3 Earth radius model as radar. Two units at 2m height have ~11 km radio horizon. One at 2m, one at 30m: ~28 km. Mountain-top relay extends dramatically. Model: reuse radar_horizon() for comms
Altitude → comms range	NOT IMPLEMENTED	No altitude benefit for comms	Higher antenna = longer radio range. Aircraft-to-ground comms have much longer range than ground-to-ground. Should improve comms reliability for airborne C2 nodes
Tropospheric ducting for comms	NOT IMPLEMENTED	Only radar ducting modeled	Same physics extends UHF/VHF range in certain weather (warm, humid, maritime). Common in Gulf operations. Model: if ducting_possible, multiply comms_range × duct_extension_factor

---

1.12 Air Combat Environment Interaction

Parameter	Status	Current State	What Should Happen
Weather sortie abort	WIRED	Global weather modifier < 0.3 = abort	—
Visibility → WVR Pk	WIRED	< 10 km visibility degrades guns/WVR Pk	—
Cloud ceiling	COMPUTED/UNCONSUMED	ConditionsEngine.air().cloud_ceiling available	Should gate: CAS operations (need visual contact with ground target), dive bombing (need to see target from altitude), low-level operations (ceiling < 500m forces low-level flight → increased ground fire risk)
Icing risk → aircraft performance	COMPUTED/UNCONSUMED	ConditionsEngine.air().icing_risk calculated	Wing ice: +15% stall speed, −20% max lift. Engine ice: −10–30% power. Radar dome ice: −3 dB signal loss. Model: if icing_risk > threshold, apply performance penalties + periodic de-ice check
Density altitude → performance	COMPUTED/UNCONSUMED	Density altitude calculated	Thin air = longer takeoff, reduced climb rate, lower max speed, reduced helicopter hover ceiling. Model: performance × (ρ/ρ₀) for thrust/lift-dependent parameters
Wind → BVR range	NOT IMPLEMENTED	BVR missile range ignores wind	Headwind reduces effective missile range; tailwind extends it. Effect is ~10–15% for strong winds. Model: effective_range = base_range × (1 ± wind_component/missile_speed)
Altitude → energy advantage	NOT IMPLEMENTED	Altitude tracked but no energy trading	Higher aircraft has more potential energy → can convert to speed in dive. Lower aircraft must climb (losing speed) to engage. This is fundamental to air combat. Model: energy_state = altitude × g + 0.5 × v² → energy advantage modifier
Turbulence → gun accuracy	NOT IMPLEMENTED	Smooth flight assumed	Turbulence degrades gun tracking accuracy. Model: guns_Pk × (1 − turbulence_factor) where turbulence = f(terrain_roughness, wind_speed, altitude_AGL)

---

1.13 Detection Environment Integration Gaps

Parameter	Status	Current State	What Should Happen
Visibility → visual detection	WIRED	Beer-Lambert atmospheric extinction (3.0/visibility)	—
Illumination → visual detection	WIRED	Lux multiplier on visual signature	—
Thermal contrast → thermal detection	WIRED (indirect)	Uses illumination-based modifier, not physics ΔT	Should use: background_temperature from TimeOfDayEngine vs target thermal signature. ΔT drives thermal detection range. At thermal crossover (dawn/dusk), ΔT → 0 → thermal detection collapses
Obscurant opacity → detection	NOT WIRED	ObscurantsEngine has opacity_at(pos) but detection never calls it	Should query: visual_opacity and thermal_opacity at target position. Smoke blocks visual (0.9), partially blocks thermal (0.1 standard, 0.8 multispectral), doesn't block radar (0.0)
Vegetation density → concealment	NOT WIRED	SeasonsEngine computes, detection ignores	Summer foliage provides concealment (+30% visual signature reduction in forest). Winter bare trees provide much less (+5%). Model: concealment_modifier = base_concealment × vegetation_density
Rain → radar detection	NOT WIRED	_compute_rain_detection_factor() exists in battle.py but never called	Rain creates radar clutter that masks targets. Already implemented with ITU-R P.838 model. Just needs call site in detection flow
Convergence zones → sonar	NOT WIRED	convergence_zone_ranges() computed	Should create detection "rings" at 55km intervals. Target between CZ ranges is in acoustic shadow. Within CZ range, detection probability spikes. Fundamental to deep-water ASW
Thermocline → sonar	NOT WIRED	Thermocline depth computed	Submarine below thermocline is very hard to detect from surface sonar (+20 dB loss). Surface ship sonar can't reliably detect deep targets. Model: if target_depth > thermocline_depth, add layer_loss to transmission
Sea state → radar clutter	NOT IMPLEMENTED	No sea surface clutter model	Radar returns from waves mask surface targets (sea clutter). Higher sea state = worse clutter. Model: clutter_noise ∝ σ₀(beaufort) × resolution_cell_area. Affects both air defense (low-flying targets) and naval radar
NVG → night detection	COMPUTED/UNCONSUMED	nvg_effectiveness() computed	NVG-equipped units should have enhanced night detection (visual modifier recovery from 0.2 to ~0.6). Currently only affects movement speed (0.7x recovery), not detection range

---

1.14 Logistics & Transport Environment Interaction

Parameter	Status	Current State	What Should Happen
Mud speed fraction	DEFINED/UNCALLED	transport.py has mud_speed_fraction=0.5	Never applied in battle-loop movement. Should slow logistics convoys by 50% in muddy conditions. Currently only MovementEngine queries ground_trafficability, not transport
Snow speed fraction	DEFINED/UNCALLED	transport.py has snow_speed_fraction=0.7	Same as mud — defined but never applied
Airlift weather gate	WIRED	Ceiling < 500m cancels airlift	—
Road condition degradation	NOT IMPLEMENTED	Roads have fixed speed bonus	Heavy rain/freeze-thaw should degrade unpaved roads. Paved roads unaffected by weather but blocked by snow/ice until cleared. Model: road_speed = base_speed × weather_condition_factor
Supply route terrain effects	NOT IMPLEMENTED	Supply routing uses networkx shortest path with fixed costs	Route costs should reflect current terrain conditions (flooded bridge, muddy road, snow-blocked pass). Dynamic edge costs from SeasonsEngine
Aircraft wind limits for cargo	NOT IMPLEMENTED	No crosswind gate for fixed-wing	Crosswind > 15 knots limits transport aircraft operations. Model: abort if crosswind > aircraft_crosswind_limit

---

1.15 Scenario Environment Configuration Gaps

Parameter	Status	What Should Be Configurable
Starting weather state	PARTIALLY WIRED	weather_conditions.precipitation maps to state. Should also allow explicit weather_state: FOG
Starting time of day	WIRED	start_time in scenario YAML sets datetime
Starting season	NOT CONFIGURABLE	Season derived from date, but SeasonsEngine ground state starts fresh (no prior accumulation)
Latitude for astronomy	PARTIALLY WIRED	Derived from scenario map center
Pre-placed smoke/fog	NOT IMPLEMENTED	No scenario YAML field
Pre-placed minefields	NOT IMPLEMENTED	No scenario YAML field
Sea state initialization	NOT WIRED	sea_state field in some scenario YAMLs but not passed to SeaStateEngine
Terrain condition overrides	NOT IMPLEMENTED	All terrain starts at default

---

Theme 2: Core Combat Integration (Priority: Highest — P0 Bugs)

These are not missing features — they are bugs. Existing code computes results that are silently discarded, engagement types are declared but never routed, and fundamental logistics gates are absent.

2.1 Air Combat Routing — 3 Engines With Wrong Physics

Problem: Air-to-air, air-to-ground, and SAM engagements all route through the generic DIRECT_FIRE path, which applies ground combat physics (range-dependent Pk with terrain modifiers). Three purpose-built engines (AirCombatEngine, AirGroundEngine, AirDefenseEngine) with correct domain physics exist but are unreachable because _infer_engagement_type() never assigns AIR_TO_AIR, AIR_TO_GROUND, or SAM.

Fix: Update _infer_engagement_type() to detect attacker/target domain combinations and assign the correct EngagementType. Route each type to its respective engine. This is routing logic only — the engines are fully implemented.

Impact: Every air engagement in every scenario changes from ground-combat Pk to proper air combat Pk. Expect significant recalibration.

2.2 Damage Detail — Computed Then Discarded

Problem: DamageEngine.resolve_damage() returns a DamageResult with 6 fields. Only damage_fraction is read — everything else (casualties, systems_damaged, fire_started, ammo_cookoff, penetrated) is discarded. Units are binary: 100% combat power until destroyed/disabled.

Fix: Extract and apply each field. Casualties reduce personnel count. Systems_damaged degrade specific equipment. Fire_started triggers fire zone creation. Ammo_cookoff applies secondary explosion to nearby units. Penetrated degrades armor for subsequent hits.

Impact: Graduated degradation replaces binary threshold. A tank with damaged optics has reduced detection. A unit with 30% casualties has reduced firepower. Fires create persistent battlefield obstacles.

2.3 Posture Protection — DUG_IN Does Nothing

Problem: DUG_IN units take the same casualties as units standing in the open. The only effect is speed=0 (they don't move). Terrain cover provides some protection, but posture itself provides zero damage reduction.

Fix: Apply posture damage reduction multiplier: DEFENSIVE ~20%, DUG_IN ~50%, FORTIFIED ~70%. These map to hasty fighting position, prepared position, and hardened position respectively. Values configurable via CalibrationSchema.

Impact: Defensive operations become viable. Currently the only winning strategy is attack (defender has no defensive advantage beyond not moving).

2.4 Logistics Gates — Units Fight Without Resources

Problem: Units with 0 fuel still move. Units with 0 ammo still fire. Movement doesn't consume fuel. Logistics is tracked and reported but never constrains combat.

Fix: Check ammo > 0 before weapon.fire(). Check fuel > 0 before movement execution. Consume fuel proportional to distance × consumption_rate during movement.

Impact: Logistics becomes consequential. Cutting supply lines degrades enemy combat power. Long advances require fuel resupply. Ammo conservation becomes a real concern.

(All Phase 58 deliverables)

---

Theme 3: C2 Friction (Priority: High)

Three C2 engines (~1,400 lines total) are fully implemented but disconnected. Wiring them creates realistic command friction — orders take time, get misunderstood, and planning delays create windows of vulnerability.

3.1 Order Propagation — Delay & Misinterpretation

Current state: OrderPropagationEngine (315 lines) computes: - Propagation delay: base_time(echelon) × type_mult × priority_mult × staff_mult + lognormal_noise - Platoon: ~5 min base, Division: ~2 hr base - FRAGO: 0.33x, WARNO: 0.1x, OPORD: 1.0x - FLASH priority: 0.1x, ROUTINE: 1.0x - Misinterpretation probability: base × (1 − staff_eff) × (1 − comms_quality) - Comms availability check (can order physically reach recipient?)

Wiring needed: - When AI commander issues orders (OODA DECIDE phase), route through OrderPropagationEngine instead of instant execution. - Delayed orders queue with scheduled delivery time. - On delivery, roll for misinterpretation — misunderstood orders execute with modified parameters (wrong objective, wrong timing). - If comms are down (comms engine reports no path), order fails entirely.

Impact: Creates Clausewitzian friction. Surprise attacks succeed because defenders can't reorganize quickly. Comms degradation (EW jamming, courier intercept) becomes tactically meaningful.

Design consideration: Must be opt-in per scenario (enable_order_propagation: true) to avoid breaking existing scenarios that assume instant C2.

3.2 Planning Process — MDMP Delays

Current state: PlanningProcessEngine (564 lines) implements: - Planning phases: RECEIVING_MISSION → ANALYZING → DEVELOPING_COA → COMPARING → APPROVING → ISSUING_ORDERS - Method selection: INTUITIVE (fast, low echelon) vs MDMP (slow, thorough, high echelon) - 1/3-2/3 rule: Commander uses 1/3 of available time for planning, subordinates get 2/3 - Phase durations scaled by method speed multipliers

Wiring needed: - When OODA cycle reaches DECIDE, check if a planning process is already underway. - If not, initiate planning with duration based on echelon and method. - Commander cannot issue new orders until planning completes. - Higher-quality planning (MDMP) produces better COA evaluation scores.

Impact: Creates realistic planning tempo. Low-echelon units (company) can react in minutes. Division-level changes take hours. OODA speed becomes a real differentiator between doctrinal schools (Maneuver warfare has faster planning loops).

Design consideration: Planning delays interact with order propagation delays — total C2 latency = planning time + propagation time. Must not create unplayable stalls for the AI.

3.3 Air Tasking Order Management

Current state: ATOPlanningEngine (459 lines) implements: - Aircraft registration with availability tracking (mission-capable, sortie limits, turnaround times) - ATO generation from CAS/strike requests with priority ordering - Sortie counting and reserve management - Publishes ATOGeneratedEvent

Wiring needed: - Register all aerial units with ATOPlanningEngine at scenario start. - CAS requests from ground commanders route through ATO queue. - Air missions allocated from available aircraft pool (not unlimited). - Turnaround time enforced between sorties. - ATO published periodically (default every 12 hours in modern era, shorter for close air support).

Impact: Air support becomes a finite resource. Multiple requests compete for limited aircraft. Attrition of aircraft reduces future sortie capacity. Creates realistic air campaign pacing.

---

Theme 4: Stratagem & Unconventional Completion (Priority: High)

4.1 Stratagem Activation

Current state: StratagemEngine (416 lines) evaluates eligibility for 9 stratagem types but never activates them. evaluate_concentration_opportunity() and evaluate_deception_opportunity() are called in battle.py but activate_stratagem() is not.

Wiring needed: - When eligibility evaluation passes, create a stratagem plan and activate it. - Each stratagem type produces effects: - CONCENTRATION: +8% force effectiveness at Schwerpunkt - DECEPTION: Enemy AI receives false force disposition estimates - FEINT: Enemy AI commits reserves to feint axis - ECONOMY_OF_FORCE: Reduced force can hold secondary sector - SURPRISE: First-engagement bonus (defender has no prepared positions) - Effects modulate combat modifiers for the duration of the stratagem.

Impact: AI commanders can employ stratagems as force multipliers. Concentration at the decisive point becomes a real mechanic (Austerlitz, 73 Easting envelopment).

4.2 Unconventional Warfare Engine

Current state: UnconventionalWarfareEngine (397 lines) implements IED, guerrilla tactics, and human shields. Never called.

Wiring needed: - Route INSURGENT/MILITIA unit engagements through unconventional engine. - IED encounters triggered by movement through insurgent-controlled areas. - Guerrilla hit-and-run: attack then disengage before response. - Human shield mechanics: reduce casualty Pk when civilian population present.

Impact: Hybrid Gray Zone scenario becomes meaningful. Insurgency dynamics (Phase 24e) can produce IED events. Afghan/Iraq-style asymmetric warfare.

4.3 Mine Warfare Completion

Current state: MineWarfareEngine (541 lines) resolves mine encounters but cannot lay mines. Ships can hit pre-placed mines but nobody places them.

Wiring needed: - lay_mines() callable by naval units with mine-laying capability. - Minefield persistence on the map as hazard zones. - Mine sweeping operations by minesweeper units. - Scenario YAML support for pre-placed minefields.

Impact: Naval blockade scenarios gain teeth. Suez Canal, Persian Gulf, WW2 Channel/Baltic mine warfare.

---

Theme 5: Space & EW Sub-Engine Activation (Priority: Medium)

Five space engines (~1,400 lines) and two EW engines (~760 lines) are instantiated but never called. These are complete implementations with sophisticated physics models.

5.1 Space ISR — Satellite Reconnaissance

SpaceISREngine (206 lines): Checks satellite overpasses, resolution-dependent target detection (vehicle <0.5m, company <5m, battalion <15m), cloud cover blocks optical but not SAR.

Wiring: Call check_overpass() each tick. On overpass, generate ISR reports that feed into FogOfWarManager (reveal enemy formations to the side with satellite access). Cloud cover from WeatherEngine gates optical sensors.

Impact: Strategic-level intelligence. Side with satellite advantage sees enemy disposition updates every ~90 minutes. GPS-denied environments (EW jamming) degrade positioning.

5.2 Early Warning — BMD Cueing

EarlyWarningEngine (143 lines): Detects ballistic missile launches via GEO/HEO early warning satellites. 30-90s detection delay, computes usable warning time.

Wiring: When a ballistic missile is fired (engagement type BALLISTIC_MISSILE), check if early warning satellites detect the launch. If detected, publish warning event that can trigger air defense interceptors.

Impact: Nuclear/ballistic missile scenarios gain defensive layer. Korean Peninsula (THAAD defense) and space scenarios become more realistic.

5.3 ASAT — Anti-Satellite Warfare

ASATEngine (353 lines): 4 ASAT weapon types (kinetic kill, co-orbital, ground laser dazzle, ground laser destruct). Debris cascade (Kessler syndrome). Kinetic Pk model.

Wiring: Route ASAT engagements through ASATEngine. Destroyed satellites removed from ConstellationManager (affects GPS, ISR, SATCOM, early warning). Debris cascade can degrade entire orbital bands.

Impact: Space warfare becomes consequential. Destroying enemy GPS constellation degrades their precision weapons. Kessler cascade can deny LEO to both sides.

5.4 SIGINT — Signals Intelligence

SIGINTEngine (488 lines): ELINT (radar parameter capture), COMINT (comms intercept), traffic analysis. AOA/TDOA geolocation. Intercept probability from SNR.

Wiring: Call attempt_intercept() when enemy units emit (radar, comms). Successful intercepts produce geolocation reports feeding FogOfWarManager. Traffic analysis reveals activity levels (enemy massing forces detectable via comms surge).

Impact: EW-capable forces can passively detect enemy radars and comms. SEAD targeting improves (find the SAM radar, then shoot HARM at it). Intelligence fusion between SIGINT + Space ISR + ground sensors.

5.5 ECCM — Electronic Protection

ECCMEngine (270 lines): 4 ECCM techniques (frequency hop, spread spectrum, sidelobe blanking, adaptive nulling) with J/S reduction calculations.

Wiring: When EW jamming is computed, query ECCMEngine for target's ECCM suite. Subtract ECCM reduction from jammer J/S ratio. Units with good ECCM resist jamming better.

Impact: Creates asymmetry between advanced forces (NATO ECCM vs Soviet-era radar) and legacy systems. EW scenarios become more nuanced (not just "jammer wins").

---

Theme 6: Cross-Module Feedback Loops (Priority: High)

These are the systemic integration gaps where one module's output should drive another module's behavior but currently doesn't. The build-then-defer pattern created modules that publish state changes as events but no module subscribes to act on them.

6.1 Detection → AI Assessment

Problem: AI assessment receives enemy_unit_count (raw ground truth) instead of querying FogOfWarManager for detected contacts with confidence levels. A side with zero sensors sees enemies perfectly.

Fix: When FOW is enabled, AI assessment reads fow_manager.get_contacts(side) — gets detected count, estimated strength, confidence. Assessment quality degrades with poor sensor coverage (fewer contacts, lower confidence → worse decisions). When FOW is disabled, behavior unchanged (backward compatible).

6.2 Medical → Unit Strength

Problem: MedicalEngine publishes CasualtyTreatedEvent and ReturnToDutyEvent but nobody subscribes. Treated casualties never return to duty — units stay permanently depleted.

Fix: Subscribe to RTD events. On receipt, increment unit personnel count (capped at original strength). Medical capacity (M/M/c queue) gates throughput — more medical assets = faster RTD.

6.3 Maintenance → Readiness

Problem: MaintenanceEngine publishes EquipmentBreakdownEvent and MaintenanceCompletedEvent. Nobody subscribes. Broken equipment continues functioning.

Fix: Subscribe to breakdown events. Reduce unit weapon/sensor count based on equipment lost. Subscribe to maintenance completed events. Restore capability. Units with >30% equipment broken marked DEGRADED (reduced Pk, reduced movement).

6.4 Checkpoint State Registration

Problem: CheckpointManager.register() is never called in production. Checkpoints save only clock + RNG state. Restoring a checkpoint produces correct time with wrong morale, wrong detection tracks, wrong supply levels, wrong equipment condition.

Fix: Register all stateful modules: morale state, detection tracks, supply levels, equipment condition, weather state, OODA phase, escalation level. Verify round-trip: save → restore → continue produces identical outcomes.

6.5 MISSILE Engagement Routing

Problem: MISSILE EngagementType is declared but never assigned by _infer_engagement_type(). MissileEngine computes flight phases and terminal guidance but is never invoked. Missiles route as DIRECT_FIRE (instant hit, wrong physics).

Fix: Assign MISSILE type for guided missile engagements (ATGM, HARM, cruise missile). Route through MissileEngine. Wire MissileDefenseEngine for intercept attempts. Flight time creates engagement delay (realistic missile time-of-flight).

6.6 Comms → C2 Degradation

Problem: Comms degradation (from EW jamming, terrain blockage, range limits) affects assessment confidence score but doesn't constrain what units can do. A unit with zero comms fights identically to one with perfect comms.

Fix: Below c2_min_effectiveness threshold, unit reverts to last received orders. No new targeting assignments, no posture changes, no disengagement. Creates realistic "comms loss = fight in place" behavior. Feeds Phase 64's order propagation failure probability.

(All Phase 63 deliverables)

---

Theme 7: Code Cleanup & Dead Field Resolution (Priority: Low)

7.1 ConditionsEngine Disposition

Current state: 249-line facade that aggregates all environment sub-engines into domain-specific query methods (land(), air(), maritime(), acoustic(), electromagnetic()). Never instantiated.

Recommendation: Instantiate as optional convenience facade. Don't refactor existing code. New Block 7 wiring can use it where aggregated queries are cleaner.

7.2 Dead YAML Field Resolution

Field	Action	Rationale
weight_kg	Keep, document as "data-only"	Useful for scenario documentation; future weight-of-fire calculations
propulsion	Wire to drag model	Rocket/turbojet/ramjet propulsion types should affect missile kinematics and altitude performance
unit_cost_factor	Keep, document as "data-only"	Future logistics cost modeling
data_link_range	Wire to C2	UAV data link range should gate C2 effectiveness beyond range

7.3 SimulationContext Stubs

7.4 P4 Dead Code Removal

Remove computed values with no meaningful consumer to clean up false audit findings: - TimeOfDayEngine.shadow_azimuth — no consumer, not useful at simulation scale - TimeOfDayEngine solar/lunar contribution decomposition — diagnostic data, only total lux matters - UnderwaterAcoustics.deep_channel_depth — SOFAR channel is extremely niche (fixed sonobuoy networks only)

(All Phase 66 deliverables)

Remove TODO comments for engines that are instantiated (SeasonsEngine, ObscurantsEngine, ConditionsEngine).

---

Comprehensive Inventory: Computed But Unconsumed Parameters

This is the definitive list of values that environment engines calculate every tick but that no downstream system reads. Each of these represents wasted computation and missing fidelity.

Engine	Parameter	Consumers That Should Use It
SeasonsEngine	mud_depth	MovementEngine (wheeled penalty), logistics transport
SeasonsEngine	snow_depth	MovementEngine (all-unit penalty), infrastructure (road degradation)
SeasonsEngine	vegetation_density	DetectionEngine (seasonal concealment), fire spread probability
SeasonsEngine	vegetation_moisture	Fire zone creation (wet = no ignition), CBRN agent persistence
SeasonsEngine	sea_ice_thickness	MovementEngine (ice crossing), naval movement (ice navigation)
SeasonsEngine	wildfire_risk	Fire zone system (spontaneous ignition probability)
SeasonsEngine	daylight_hours	Operational planning, fatigue rate
SeaStateEngine	wave_period	Ship resonance roll, helicopter deck landing
SeaStateEngine	tidal_current_speed	Ship movement, mine drift, submarine station-keeping
SeaStateEngine	tidal_current_direction	Same as speed
SeaStateEngine	beaufort_scale	Small craft operations gate, deck operations gate, sea clutter
UnderwaterAcoustics	surface_duct_depth	Sonar detection (in-duct enhancement, below-duct shadow)
UnderwaterAcoustics	thermocline_depth	Sonar detection (+20 dB loss for targets below layer)
UnderwaterAcoustics	deep_channel_depth	SOFAR channel long-range detection
UnderwaterAcoustics	convergence_zone_ranges	Sonar detection rings at 55km intervals
EMPropagation	radar_horizon()	Air defense detection range gate (Earth curvature)
EMPropagation	free_space_path_loss()	Comms range, radar range, EW intercept range
EMPropagation	atmospheric_attenuation()	Radar/comms degradation in rain/humidity
EMPropagation	ducting_possible + duct_height	Radar/comms range extension in warm humid maritime
EMPropagation	hf_propagation_quality()	HF radio reliability (day/night D-layer/F-layer)
WeatherEngine	wind.gust	Helicopter ops, bridge-laying, parachute drops, amphibious small craft
WeatherEngine	pressure	Air density (ideal gas), altimeter accuracy
TimeOfDay	background_temperature	Thermal detection SNR (target vs background ΔT)
TimeOfDay	crossover_in_hours	Thermal detection vulnerability window (ΔT → 0)
TimeOfDay	shadow_azimuth	Visual detection angle dependency
TimeOfDay	nvg_effectiveness()	Night detection range for NVG-equipped units
Terrain	vegetation_height	LOS check (ground-level vegetation blocks), helicopter clearance
Terrain	combustibility	Fire zone ignition probability
Obstacles	traversal_risk	Casualty probability during crossing
Obstacles	traversal_time_multiplier	Movement delay through obstacles
Hydrography	is_fordable(), ford_points_near()	River crossing at ford points with time cost
Infrastructure	Bridge.capacity_tons	Heavy vehicle weight gate
Infrastructure	Road.speed_factor override	Per-road speed instead of hardcoded table
Equipment	temperature_range + environment_stress()	Weapon jam rate, electronics failure rate
Battle.py	_compute_rain_detection_factor()	Radar detection probability in rain (ITU-R P.838) — function exists, never called
DEW Engine	humidity/precip_rate parameters	Atmospheric transmittance — accepts params, battle loop never passes them

Total: 36 unconsumed parameters across 11 engines. ~4,500 lines of computation producing values nobody reads.

---

Proposed Phase Structure

Phase 58: Structural Verification & Core Combat Wiring

Focus: Create structural verification tests that prevent future regression, then fix all P0 combat integration bugs — air combat routing, damage detail extraction, posture protection, and logistics gates. These are the highest-impact items: existing code that should work but doesn't.

Why first: Structural tests must exist before wiring begins so that every subsequent phase runs against automated audits. P0 combat items affect every engagement and are the most impactful single changes in this block.

Deliverables:

Structural Verification Tests (run continuously from here forward): - Unconsumed parameter audit: Instrument engine classes, run representative scenario, assert no property written but never read (allowlist for P4 items) - Dead method audit: AST-parse all source files, verify every public engine method has at least one external caller (beyond tests) - Event subscription audit: Verify every published event type has at least one functional subscriber (beyond Recorder) or is in OBSERVATION_ONLY allowlist - Engagement routing completeness: Every EngagementType enum value is assignable by _infer_engagement_type() and has a resolution handler - Feedback loop verification: Depleted ammo prevents firing, depleted fuel prevents movement, DUG_IN reduces damage, checkpoint round-trips all state

Air Combat Routing (P0): - Wire AIR_TO_AIR EngagementType → AirCombatEngine (BVR + WVR physics, not ground combat Pk) - Wire AIR_TO_GROUND EngagementType → AirGroundEngine (CAS/strike Pk model with altitude, dive angle, CEP) - Wire SAM EngagementType → AirDefenseEngine (missile Pk with engagement envelope, ECM effects) - Update _infer_engagement_type() to assign these types based on attacker/target domain (AIR vs GROUND/NAVAL)

Damage Detail Extraction (P0): - Extract DamageResult.casualties → reduce unit personnel count (not just damage_fraction threshold) - Extract DamageResult.systems_damaged → degrade specific equipment on unit (weapon destroyed, sensor damaged) - Extract DamageResult.fire_started → trigger fire zone creation at target position (feeds Phase 60 fire system) - Extract DamageResult.ammo_cookoff → secondary explosion damage to nearby units - Extract DamageResult.penetrated → armor degradation (reduced protection on subsequent hits)

Posture Protection (P0): - DEFENSIVE → ~20% damage reduction (hasty fighting position) - DUG_IN → ~50% damage reduction (prepared position) - FORTIFIED → ~70% damage reduction (hardened position) - Protection applied as multiplier on incoming damage_fraction before threshold comparison - CalibrationSchema fields for posture protection values (per-scenario tuning)

Logistics Gates (P0): - Ammo depletion → weapon cannot fire (check remaining rounds before weapon.fire()) - Fuel depletion → unit cannot move (check fuel > 0 before movement execution) - Fuel consumption by movement (distance × consumption_rate → fuel drawdown)

Tests: ~55-65

Phase 59: Atmospheric & Ground Environment Wiring

Focus: Wire all computed-but-unconsumed atmospheric and ground parameters. Make seasons, weather, and terrain fully contribute to movement, detection, and combat.

Deliverables: - Seasons → Movement: Mud depth penalty (wheeled vs tracked differentiation), snow depth penalty, ice crossing (weight-dependent), vegetation speed penalty through dense terrain - Seasons → Detection: Vegetation density as seasonal concealment modifier, vegetation height for ground-level LOS - Seasons → Infrastructure: Snow depth degrades road speed bonus, frozen water bodies traversable - Weather → Ballistics: Pressure + humidity → true air density (ideal gas law correction), temperature lapse rate at altitude - Weather → Operations: Wind gust threshold for helicopter/parachute/bridge-laying abort, high wind halt for infantry - Equipment → Reliability: Wire temperature_range stress → weapon jam rate, electronics failure - Terrain → Movement: Wire obstacle traversal_risk and traversal_time, ford crossing with depth/current, bridge capacity gate - Propellant temperature: MV modifier from ambient temperature (MIL-STD-1474)

Tests: ~45-55

Phase 60: Obscurants, Fire, & Visual Environment

Focus: Instantiate and wire ObscurantsEngine. Create fire zone system linked to Phase 58's damage detail extraction (fire_started). Wire all visual/thermal environment parameters.

Deliverables: - ObscurantsEngine instantiation: Create on SimulationContext, call update() each tick - Smoke from combat: Artillery/mortar impacts spawn smoke clouds, drift with wind, decay (30min half-life) - Dust from movement: Vehicles on dry terrain spawn dust trails (signature enhancement + LOS degradation) - Fog auto-generation: Weather FOG state triggers ObscurantsEngine fog deployment - Smoke/fog → Detection: opacity_at(pos) degrades visual/thermal detection SNR per spectral blocking table - Smoke/fog → Combat: Obscurant between attacker/target reduces Pk for visual/thermal-guided weapons - Thermal environment wiring: Background temperature + target temp → true ΔT for thermal detection. Crossover vulnerability windows - NVG → detection: NVG-equipped units get night detection range recovery (not just movement) - Fire zones: Wire IncendiaryDamageEngine (P1). fire_started from DamageResult (Phase 58) triggers fire zone creation. Fire produces smoke (feeds ObscurantsEngine). Fire blocks movement. Fire decays when fuel (vegetation) exhausted - Combustibility wiring: Terrain combustibility gates fire ignition probability - Scenario YAML: environment_config for pre-placed smoke/fog zones, season/ground state overrides - Artificial illumination: Flare events that temporarily raise illumination in area

Tests: ~45-55

Phase 61: Maritime, Acoustic, & EM Environment

Focus: Wire all maritime, underwater acoustic, and electromagnetic propagation parameters. Wire CarrierOpsEngine (P1) for naval air operations.

Deliverables: - Sea state → ship movement: Beaufort-dependent speed penalty (small craft more affected) - Sea state → carrier ops: Launch/recovery gate by Beaufort scale - CarrierOpsEngine wiring (P1): CAP station management, recovery windows, sortie tracking for carrier-based aircraft - Sea state → small craft: Amphibious landing craft casualty risk + speed penalty in high seas - Sea state → radar: Sea clutter model masking surface targets - Swell direction: Roll factor from wave direction vs ship heading - Tidal current → movement: Ship speed adjustment (with/against), mine drift, submarine station-keeping - Wave period → ship resonance: Roll amplitude when wave period matches hull natural frequency - Acoustic wiring: Surface duct detection enhancement/shadow, thermocline layer loss (+20 dB), convergence zone detection rings at 55km intervals - Radar horizon: Gate air defense detection by 4/3 Earth curvature model - EM ducting: Extend radar range in warm/humid maritime conditions - Atmospheric attenuation: Frequency-dependent rain loss for radar and comms - HF propagation: Wire hf_quality → comms reliability (day/night D-layer/F-layer) - Radio horizon: Altitude-dependent comms range using same 4/3 Earth model - DEW wiring: Pass humidity/precip_rate/dust to DEW engine (call site fix), sea spray attenuation, dust distinction from rain - Rain detection: Wire existing _compute_rain_detection_factor() call site in battle loop

Tests: ~50-60

Phase 62: Human Factors, CBRN, & Air Combat Environment

Focus: Environmental effects on personnel, CBRN-environment interaction, and air domain environmental coupling.

Deliverables: - Heat stress: WBGT model for heat casualties (temperature × humidity × MOPP level × exertion) - Cold injury: Wind chill formula + exposure time → cold casualty rate - MOPP degradation: Beyond speed — reduce FOV (detection), dexterity (reload speed), voice clarity (comms quality) - Altitude sickness: Performance degradation above 2500m - Environmental fatigue: Temperature/altitude stress accelerates fatigue accumulation - CBRN rain washout: Precipitation scavenges airborne agents (exponential washout) - CBRN temperature persistence: Arrhenius agent decay rate (nerve agent half-life 2hr at 40°C vs 24hr at 0°C) - CBRN inversion trapping: Temperature inversion detection → 5-10x concentration multiplier - CBRN UV degradation: Solar radiation breaks down agents in daylight - Air combat: cloud ceiling gate: CAS/dive bombing requires visual contact below ceiling - Air combat: icing penalties: Wing ice, engine ice, radar dome ice degradation - Air combat: density altitude: Engine thrust/lift penalties at high altitude + hot temperature - Air combat: wind → BVR range: Headwind/tailwind modifies missile effective range - Air combat: altitude energy advantage: Higher aircraft gets energy_state modifier in engagement

Tests: ~40-50

Phase 63: Cross-Module Feedback Loops

Focus: Wire all P1 cross-module integration gaps — the feedback loops where one system's output should drive another system's behavior but currently doesn't. This phase closes the systemic build-then-defer-wiring pattern.

Deliverables:

Detection → AI (P1): - AI assessment reads FOW contacts (detected count, estimated strength, confidence) instead of ground truth enemy count - When enable_fog_of_war is True, AI assessment quality degrades with sensor coverage gaps - When FOW is disabled, behavior is unchanged (backward compatible)

Medical → Strength (P1): - ReturnToDutyEvent consumption: treated casualties restore to unit personnel count - CasualtyTreatedEvent consumption: track treatment pipeline (triage → treatment → RTD) - Medical capacity limits (M/M/c queue) gate RTD throughput

Maintenance → Readiness (P1): - EquipmentBreakdownEvent consumption: broken equipment reduces unit combat power (weapon count, sensor availability) - MaintenanceCompletedEvent consumption: repaired equipment restores capability - Maintenance state feeds readiness: units with >30% equipment broken are DEGRADED (reduced Pk, reduced movement)

Checkpoint State Registration (P1): - Register all stateful modules with CheckpointManager: morale state, detection tracks, supply levels, equipment condition, weather state, OODA phase, escalation level - Checkpoint round-trip test: save → restore → continue produces identical outcomes to uninterrupted run - Prioritize modules with expensive state (detection tracks, PRNG streams, morale matrices)

MISSILE Engagement Routing (P1): - Wire MISSILE EngagementType → MissileEngine (flight phase + terminal guidance) - Wire MissileDefenseEngine → intercept attempts (launch-on-detect or launch-on-warning) - Missile flight time creates engagement delay (not instant hit like DIRECT_FIRE)

Comms → C2 (P1): - Comms loss (from EW jamming, terrain LOS, range) degrades C2 effectiveness below c2_min_effectiveness - Units with zero comms revert to last received orders (no new targeting, no posture changes) - Comms degradation feeds order propagation failure probability (Phase 64)

Tests: ~40-50

Phase 64: C2 Friction & Command Delay

Focus: Wire the three dormant C2 engines to create realistic command friction. Builds on Phase 63's comms → C2 degradation wiring.

Deliverables: - OrderPropagationEngine → OODA cycle (order delay + misinterpretation + comms check) - PlanningProcessEngine → OODA DECIDE phase (planning duration before order issue) - ATOPlanningEngine → air operations (sortie management, CAS request queue) - StratagemEngine full activation (9 stratagem types with combat effects) - C2 friction opt-in flag (enable_c2_friction: true in scenario YAML) - Environmental coupling: comms weather degradation → order propagation failure - Environmental coupling: HF day/night quality → comms reliability → C2 latency

Tests: ~35-45

Phase 65: Space & EW Sub-Engine Activation

Focus: Wire the five dormant space engines and two dormant EW engines.

Deliverables: - SpaceISREngine → FogOfWarManager (satellite ISR passes reveal formations) - EarlyWarningEngine → air defense (BMD cueing from early warning satellites) - ASATEngine → space warfare (satellite destruction, debris cascade, constellation degradation) - SIGINTEngine → FogOfWarManager (emitter geolocation, traffic analysis) - ECCMEngine → EW jamming loop (ECCM J/S reduction) - Space sub-engine delegation verified end-to-end

Tests: ~35-45

Phase 66: Unconventional, Naval, & Cleanup

Focus: Wire remaining dormant engines and clean up dead code.

Deliverables: - UnconventionalWarfareEngine → battle routing (IED, guerrilla, human shields) - MineWarfareEngine completion (mine laying, minefield persistence, sweeping, scenario YAML minefields) - SiegeEngine → campaign loop (P2 — ancient/medieval siege mechanics) - AmphibiousAssaultEngine → naval ops (P2 — beach assault state machine) - ConditionsEngine instantiation (optional facade) - Dead YAML field resolution (wire propulsion → drag model, data_link_range → C2 range gate) - P4 dead code removal (shadow_azimuth computation, solar/lunar decomposition, deep_channel_depth) - SimulationContext stub cleanup

Tests: ~30-35

Phase 67: Integration Validation & Recalibration

Focus: Validate all scenarios still produce correct outcomes with all new systems active. Run structural verification tests to confirm zero remaining gaps. This is the final hardening pass.

Deliverables: - Full scenario evaluation with all new systems active - Recalibration of affected scenarios (expect significant recalibration — air combat routing, posture protection, logistics gates, and environmental effects all change outcomes) - MC validation at 80% threshold with N=10 seeds - Structural verification test results: - Zero unconsumed parameters (beyond P4 allowlist) - Zero uncalled public engine methods (beyond test-only methods) - All event types have functional subscribers (or are in observation-only allowlist) - All EngagementType values routable and handled - All feedback loops functional (ammo→fire, fuel→move, posture→damage, checkpoint→state) - CalibrationSchema exercised fields audit (17 never-set fields either exercised in scenarios or removed) - Regression test suite for all new environmental and combat effects - Cross-doc audit and documentation sync - Block 7 postmortem

Tests: ~35-45

---

Test Count Estimate

Phase	Focus	Tests	Cumulative (Python)
58	Structural Verification & Core Combat Wiring	~60	~8,443
59	Atmospheric & Ground Environment	~50	~8,493
60	Obscurants, Fire, & Visual Environment	~50	~8,543
61	Maritime, Acoustic, & EM Environment	~55	~8,598
62	Human Factors, CBRN, & Air Combat	~45	~8,643
63	Cross-Module Feedback Loops	~45	~8,688
64	C2 Friction & Command Delay	~40	~8,728
65	Space & EW Sub-Engine Activation	~40	~8,768
66	Unconventional, Naval, & Cleanup	~30	~8,798
67	Integration Validation & Recalibration	~40	~8,838
Block 7 total		~455	~8,838

---

Design Decisions Needed

1. Opt-in vs Default for New Systems

Question: Should C2 friction, obscurants, space effects, etc. be on by default or opt-in per scenario?

Recommendation: - Default-on: All environmental effects (seasons, obscurants, sea state, EM propagation, atmospheric corrections). These are passive modifiers — when conditions are benign (clear weather, no smoke, dry ground), they produce unity multipliers and don't change outcomes. When conditions are adverse, they add realism. - Opt-in: C2 friction (enable_c2_friction: true). Dramatically changes battle pacing; existing scenarios assume instant C2. - Opt-in: Human factors casualties (heat/cold stress). Can produce unexpected casualties in historical scenarios not calibrated for them. - Gated by config presence: Space/EW sub-engines (only active when space_config or ew_config present).

2. ConditionsEngine: Wire or Remove?

Recommendation: Instantiate as optional convenience facade. Don't refactor existing code. Use for new Block 7 wiring where aggregated queries are cleaner.

3. Fire Spread Model Complexity

Options: - Simple: Fire zones persist, don't spread. Decays when fuel exhausted. - Medium: Fire spreads to adjacent vegetated cells at rate proportional to wind and vegetation density. Cellular automaton with wind bias. - Complex: Full wildfire model with firebreaks, slope effects, spotting (ember transport).

Recommendation: Start with medium — the cellular automaton with wind bias captures the key dynamic (fire moves downwind through dry vegetation) without the complexity of spotting/embers. The SeasonsEngine already computes vegetation_density and vegetation_moisture per cell; the WeatherEngine provides wind. The pieces are there.

4. Performance Budget

New environmental queries (obscurant opacity, seasonal trafficability, EM propagation, atmospheric corrections) add per-engagement computation. Current tick time is ~50-100ms.

Recommendation: Budget 30% overhead (15-30ms per tick). Profile after each phase. Key mitigations: - Environmental queries should be O(1) lookups with spatial caching (STRtree for obscurants, grid cache for seasons) - Radar horizon is constant per antenna height — cache per unit - EM ducting is constant per weather state — cache per tick - Thermal crossover timing changes slowly — cache per hour - Ice/mud/snow depth changes slowly — cache per tick

5. Backward Compatibility

All new systems must be backward-compatible. Existing scenarios that don't specify environment_config, enable_c2_friction, space_config, etc. should behave identically to Block 6.

Mechanism: Default values on all new config fields that produce unity modifiers (1.0 multiplier, 0.0 additive). New engines gated by config presence. ObscurantsEngine produces zero opacity when no smoke/dust/fog deployed. SeasonsEngine trafficability defaults to 1.0 when not explicitly set. Atmospheric corrections should produce negligible changes at sea level standard conditions.

6. Thermal Detection Model

Question: Should thermal detection switch from illumination-based modifier to physics-based ΔT model?

Recommendation: Yes. The ΔT model (target_temperature − background_temperature) is physically correct and uses values already computed by TimeOfDayEngine. The illumination-based modifier is a proxy that doesn't capture thermal crossover (dawn/dusk vulnerability windows where ΔT → 0 and thermal detection collapses). This is a real-world planning factor that should be represented.

7. Casualty Models for Environmental Effects

Question: Should heat/cold casualties use simple threshold models or continuous exposure models?

Recommendation: Continuous exposure. Heat casualties accumulate with WBGT × time × exertion_level. Cold casualties accumulate with wind_chill × exposure_time. Both produce casualties at a rate, not a threshold — this matches military medical data (heat/cold casualties are a function of cumulative exposure, not instant onset).

---

Risk Assessment

Risk	Severity	Mitigation
Air combat routing changes every air engagement	High	Phase 58 (earliest); recalibrate immediately; expect all air-heavy scenarios affected
Posture protection makes defenders much harder to destroy	High	Calibrate protection values conservatively; DUG_IN ~50% not ~90%
Logistics gates stop combat in scenarios without supply units	High	Gate only when logistics system is active (supply units present); degrade gracefully
Damage detail extraction produces unexpected cascading losses	High	Secondary effects (fire, cookoff) opt-in via CalibrationSchema; disabled by default
Environmental corrections change outcomes in existing scenarios	High	Phase 67 dedicated to recalibration; run full eval after each environmental phase
Seasonal trafficability breaks movement in all scenarios	High	Default trafficability=1.0 for benign conditions; only penalizes mud/snow/saturated
Obscurant smoke from barrages degrades all engagement Pk	Medium	Smoke only at impact area, not global; 30min half-life; spectral: radar unaffected
Thermal ΔT model changes detection balance	Medium	Calibrate ΔT ranges to match existing detection behavior at nominal conditions
C2 friction makes AI unable to function	High	Opt-in flag; careful tuning of delay durations; INTUITIVE method for low echelon
Heat/cold casualty model produces unexpected losses	Medium	Opt-in flag for human factors; threshold calibrated to historical data
Radar horizon gate makes many ground radars useless	Medium	Only applies to detecting targets below geometric horizon; most engagements within horizon
Fire spread model creates runaway terrain destruction	Medium	Spread rate capped; fire decays when fuel exhausted; limited by vegetation moisture
EM ducting extends detection unrealistically	Low	Only activates in specific weather; factor capped at 3× range extension
Space sub-engines add per-tick overhead	Low	Gate by space_config presence; skip when null
Scenario recalibration cascade across 37 scenarios	High	Recalibrate one at a time; accept new baselines; environmental realism justifies changes
Performance degradation from 36 new per-engagement queries	Medium	O(1) lookups with caching; profile after each phase; budget 30% overhead
ECCM wiring changes EW balance in gulf_war_ew_1991	Medium	Verify scenario still produces correct winner; recalibrate if needed

---

Research Sources

Atmospheric Physics

• ISA (International Standard Atmosphere): Lapse rate, tropopause, pressure-altitude relationship
• MIL-STD-1474: Propellant temperature effects on muzzle velocity
• ITU-R P.838: Specific rain attenuation model (dB/km vs rain rate and frequency)
• ITU-R P.676: Atmospheric gaseous attenuation (O₂ at 60 GHz, H₂O at 22 GHz)
• Pasquill-Gifford: Atmospheric stability classification and dispersion parameters

Thermal & Optical

• FLIR Systems: Thermal contrast and NETD (Noise Equivalent Temperature Difference) modeling
• MIL-HDBK-141: Electro-optical systems atmospheric propagation
• FM 3-11.50 / ATP 3-11.50: Smoke/obscurant employment doctrine
• Mie scattering theory: Particulate (dust/fog/smoke) attenuation of optical/IR beams

Maritime & Acoustic

• Pierson-Moskowitz: Wave spectrum (already implemented)
• Mackenzie equation: Sound velocity profile (already implemented)
• Urick: Underwater acoustic propagation — convergence zones, shadow zones, surface duct
• NATO STANAG 1008: Sea state definitions and operational limits
• Beaufort scale: Operational limits for small craft, carrier operations, helicopter deck landing

Electromagnetic Propagation

• ITU-R P.453: Radio refractive index of the atmosphere (ducting conditions)
• ITU-R P.526: Propagation by diffraction (terrain LOS)
• Bean & Dutton: Radio refractivity (4/3 Earth model, super-refraction)

Terrain & Ground

• US Army FM 5-33: Terrain Analysis — trafficability by soil type and moisture
• NATO AEP-55: Vehicle mobility in snow, mud, sand
• Ice thickness bearing capacity: Gold formula for ice crossing loads

Human Factors

• TB MED 507: Heat stress control and heat casualty management (WBGT model)
• AR 40-501 / FM 4-25.12: Cold injury prevention (wind chill, frostbite times)
• IAM Report 313: Altitude effects on human performance

C2 Friction

• Clausewitz: Fog of war and friction as fundamental warfare concepts
• Boyd: OODA loop timing as competitive advantage
• US Army FM 5-0: The Operations Process — MDMP phases and timing
• van Creveld: Command in War — communications delay as historical constant

Space Warfare

• Kessler & Cour-Palais (1978): Collision frequency and cascade model
• DoD Space Policy: Orbital regime definitions and satellite vulnerability
• SBIRS/DSP: Early warning satellite detection timelines

CBRN-Environment Interaction

• FM 3-11: Chemical agent persistence tables (temperature/humidity dependent)
• Seinfeld & Pandis: Atmospheric Chemistry — washout coefficients for particulate scavenging
• Arrhenius kinetics: Temperature-dependent agent degradation rates

---

Dependencies on Block 8 (Scenario Expansion)

Block 7 creates the infrastructure that Block 8's scenarios will exercise:

Block 7 Feature	Block 8 Scenario Need
Seasonal trafficability (mud/snow/ice)	Eastern Front (rasputitsa), Napoleonic Russia, Korean Winter, Finnish Winter War, Bulge
Smoke/obscurants	D-Day smoke screens, WW1 gas+smoke, Napoleonic battery smoke, 73 Easting dust
Dust from movement	North Africa (El Alamein), Gulf War desert, 73 Easting, Golan Heights
Fire zones + spread	Tokyo firebombing, Dresden, incendiary tactics, scorched earth
Sea state → operations	Falklands storms, D-Day Channel weather, Midway, Jutland
Radar horizon + EM ducting	Maritime radar scenarios, Persian Gulf ducting, air defense engagement ranges
Thermal ΔT + crossover	Dawn/dusk attack scenarios (Golan 1973, Yom Kippur), night operations
HF propagation day/night	WW2 naval (night comms skip), Atlantic convoy
Rain → radar + DEW	Gulf War (clear weather advantage), Falklands (frequent rain)
Heat/cold casualties	Chosin Reservoir, Stalingrad winter, North Africa summer, Vietnam humidity
CBRN rain washout + persistence	Halabja (hot, dry = persistent), WW1 gas (rain washout), Cold War scenarios
Acoustic layers (thermocline/CZ)	Atlantic ASW, Pacific submarine warfare, Falklands naval
C2 friction (order delays)	Any multi-echelon scenario (Kursk, Normandy, Bulge, Waterloo)
ATO sortie management	Air-centric scenarios (Bekaa Valley, Gulf War air campaign)
Stratagems	Austerlitz (concentration + deception), 73 Easting (envelopment), Cannae
Space ISR/ASAT	Modern near-peer (Taiwan, Korea, Suwalki with space denial)
SIGINT/ECCM	Cold War EW scenarios, Gulf War SEAD, Bekaa Valley
Mine warfare	Dardanelles, Persian Gulf, Baltic WW2, Falklands
Unconventional	Afghanistan, Iraq, Vietnam, Peninsular War guerrilla
Bridge capacity + fording	Rhine crossing, Arnhem, Remagen, Napoleonic river crossings
Ice crossing	Eastern Front (Don, Volga), Finnish Winter War, Chosin Reservoir
Density altitude	Afghan mountain warfare, Chosin Reservoir, Andes
Fog scenarios	Austerlitz morning fog, Falklands sea fog, English Channel
Air combat routing (proper Pk)	Every air-heavy scenario: Bekaa Valley, Gulf War, Korea, Taiwan, Midway, Battle of Britain
Posture protection (DUG_IN/FORTIFIED)	Every defensive scenario: Kursk, Stalingrad, Chosin, Verdun, trench warfare, sieges
Damage degradation (partial losses)	All scenarios benefit from graduated degradation vs binary destroy/disable
Logistics gates (fuel/ammo)	Operational-scale: Bulge (fuel crisis), North Africa (supply lines), Barbarossa (overextension)
Detection → FOW-based AI	Near-peer with contested sensor coverage: Taiwan, Suwalki, Korea, naval scenarios
Medical RTD + maintenance	Attrition scenarios: Stalingrad, Verdun, WW1 trench, Korean War
Checkpoint state	Any scenario requiring save/restore (campaign-length scenarios, multi-day operations)
MISSILE routing	Any guided missile scenario: Gulf War HARM, Taiwan anti-ship, Korean BMD