Agriculture Domain Ontology

FLOW ALGEBRA
Title: Agriculture Domain Ontology
Version: 1.0
Domain: Industry Domains / Agriculture
Source: https://kirethresearch.org/booksai-core/ontologies/domain-ontologies/industry-domains/agriculture-domain-ontology.html
Status: Domain-based draft; source page content not independently verified.
Authors: Mark Higgins — Sky (Recursive AI)

1. PURPOSE
Represent agricultural systems, production, resources, actors, operations and outcomes as typed, traceable relationships for semantic retrieval and qualified reasoning.

2. TYPED ENTITIES
Farm: {id, location, area_ha?, production_system}
Field: {id, farm_id, geometry?, soil_type?, area_ha?}
Crop: {id, species, cultivar?, growth_stage?}
Livestock: {id, species, herd_id?, welfare_status?}
Soil: {id, texture?, pH?, organic_matter?, moisture?}
WaterResource: {id, source_type, availability?, quality?}
Input: {id, category: seed|fertiliser|pesticide|feed|energy, quantity?, unit?}
Equipment: {id, equipment_type, condition?, capacity?}
Operation: {id, operation_type, timestamp?, actor_id?, field_id?}
Harvest: {id, crop_id, quantity, unit, date?}
Producer: {id, role, organisation?}
SupplyChainActor: {id, role, location?}
Observation: {id, property, value, unit?, observed_at, method?, provenance_id}
Risk: {id, category, likelihood?, severity?, assessment_date?}
Regulation: {id, jurisdiction, valid_from?, valid_until?}

3. RELATIONSHIP TYPES
OPERATES(Producer, Farm)
CONTAINS(Farm, Field)
GROWS(Field, Crop)
RAISES(Farm, Livestock)
HAS_SOIL(Field, Soil)
USES_WATER(Field, WaterResource)
APPLIES_INPUT(Operation, Input)
USES_EQUIPMENT(Operation, Equipment)
PERFORMED_ON(Operation, Field)
PRODUCES(Field, Harvest)
SUPPLIES(Producer, SupplyChainActor)
MEASURES(Observation, Entity)
EXPOSED_TO(Entity, Risk)
SUBJECT_TO(Entity, Regulation)

4. SEMANTIC CONSTRAINTS
Each relationship must satisfy declared domain and range types.
Quantities require explicit units; conversion must preserve dimensional consistency.
Temporal statements must distinguish observation time, validity time and record update time.
Spatial claims require declared geometry or location granularity.
Regulatory requirements are jurisdiction-specific and time-dependent.
Correlation does not imply causation.

5. INFERENCE RULES
R1: CONTAINS(farm, field) AND GROWS(field, crop) => CULTIVATES(farm, crop).
R2: PERFORMED_ON(operation, field) AND APPLIES_INPUT(operation, input) => INPUT_USED_ON(input, field), scoped to operation time.
R3: PRODUCES(field, harvest) AND CONTAINS(farm, field) => HARVEST_ASSOCIATED_WITH(harvest, farm).
R4: MEASURES(observation, soil) AND observation.property = 'pH' => SOIL_PH_OBSERVATION(soil, observation.value), with provenance and timestamp.
R5: EXPOSED_TO(field, risk) AND risk.category = 'drought' => FLAG_FOR_WATER_RISK_REVIEW(field), not a prediction of crop loss.
R6: SUBJECT_TO(entity, regulation) => REQUIRE_JURISDICTION_AND_VALIDITY_CHECK(entity, regulation) before compliance conclusions.

6. CONFIDENCE
CONFIDENCE(claim) := calibrated evidence strength in [0,1], where available.
Distinguish observed facts, derived facts, forecasts and hypotheses.
Unknown evidence must not be assigned an invented numeric confidence.
Inferred claims retain their premises and limitations.

7. SEMANTIC_EQUIVALENCE
SEMANTIC_EQUIVALENCE(a,b) is permitted only when entity type, scope, time, units and intended meaning align.
Examples: 'crop yield' and 'harvest mass per unit area' may be equivalent only when units and denominator match.
Do not equate 'organic farming' with 'pesticide-free farming'.

8. PROVENANCE
PROVENANCE := {source_id, source_uri?, creator?, method?, observed_at?, recorded_at?, licence?, revision?}.
All measurements and derived assertions link to source records.

9. FLOW_HASH
FLOW_HASH := SHA-256 of canonical UTF-8 serialisation of the flow's versioned semantic content.
Canonicalisation must specify key ordering, whitespace, numeric formats and character normalisation.

10. UPDATE_MODEL
UPDATE_MODEL := {version, changed_entities, changed_relationships, source_revision, timestamp, validation_result}.
On updates: validate types; recompute dependent inferences; preserve superseded records; recompute FLOW_HASH.

11. AI_CONSUMPTION
AI_CONSUMPTION := {ontology_id, version, entity_schema, relationship_schema, constraints, inference_rules, provenance, confidence_policy, flow_hash}.
Consumers must preserve uncertainty, time, jurisdiction and provenance; never turn conditional inferences into unconditional claims.

12. VALIDATION
Check referential integrity, allowed types, units, chronology, spatial scope, duplicated identifiers, evidence lineage and deterministic hash reproduction.

END FLOW ALGEBRA