Biological Environmental Indicator Ontology

FLOW ALGEBRA
Biological Environmental Indicator Ontology — Full Flow Algebra v1.0
Kireth Semantic Library
Authors: Mark Higgins — Sky (Recursive AI)
Source: https://kirethresearch.org/booksai-core/entity-graph/biological-entities/experimental-biological-entities/biological-environmental-indicator-ontology.html
Status: Newly authored conceptual formalisation; not a verified extraction or executable implementation.
Locale: en-GB

flow "BiologicalEnvironmentalIndicatorOntology" v1.0

# 1. DOMAIN AND PURPOSE
ontology.id: biological_environmental_indicator
ontology.parent: experimental_biological_entities
ontology.purpose: represent biological observations used to infer environmental conditions, pressures and change while preserving uncertainty and provenance

# 2. TYPED ENTITIES
entity BiologicalEnvironmentalIndicator <: EcologicalEntity:
  id: URI
  name: String
  indicator_kind: Enum[species, community, functional_trait, biomarker, physiological_response, ecological_index]
  target_condition: URI
  observed_response: Measurement?
  reference_condition: URI?
  location: URI
  observation_time: DateTime
  method: URI
  confidence: Float[0..1]
  provenance: URI

entity IndicatorSpecies <: BiologicalEnvironmentalIndicator:
  taxon: URI
  abundance: NonNegativeReal?
  presence: Boolean?

entity CommunityIndicator <: BiologicalEnvironmentalIndicator:
  taxa: Set[URI]
  richness: NonNegativeInteger?
  diversity_index: Real?

entity BiomarkerIndicator <: BiologicalEnvironmentalIndicator:
  analyte: URI
  organism: URI
  concentration: Measurement?

entity EnvironmentalCondition:
  id: URI
  variable: String
  value: Measurement
  temporal_extent: Interval
  spatial_extent: URI

entity EnvironmentalPressure:
  id: URI
  pressure_kind: Enum[pollution, eutrophication, warming, acidification, salinisation, habitat_loss, oxygen_depletion, disturbance, other]
  intensity: Measurement?

entity Observation:
  id: URI
  subject: URI
  value: Measurement
  sampling_method: URI
  detection_limit: Measurement?
  observed_at: DateTime
  source: URI

entity ReferenceBaseline:
  id: URI
  habitat_type: URI
  metric: URI
  distribution: StatisticalDistribution
  validity_period: Interval
  method: URI

entity Assessment:
  id: URI
  indicator: URI
  condition: URI
  evidence: Set[URI]
  outcome: Enum[improving, stable, deteriorating, indeterminate]
  confidence: Float[0..1]

# 3. TYPED RELATIONSHIPS
relation INDICATES(BiologicalEnvironmentalIndicator, EnvironmentalCondition)
relation RESPONDS_TO(BiologicalEnvironmentalIndicator, EnvironmentalPressure)
relation OBSERVED_IN(BiologicalEnvironmentalIndicator, Habitat)
relation MEASURED_BY(BiologicalEnvironmentalIndicator, Observation)
relation COMPARED_WITH(Observation, ReferenceBaseline)
relation SUPPORTS(Observation, Assessment)
relation CONFOUNDED_BY(Assessment, EnvironmentalCondition)
relation DERIVED_FROM(Assessment, Set[Observation])
relation HAS_TAXON(IndicatorSpecies, Taxon)
relation HAS_PROVENANCE(Observation, Source)

# 4. SEMANTIC CONSTRAINTS
constraint C1: every indicator assessment has a location, time, measurement method and provenance
constraint C2: confidence in [0,1]
constraint C3: baseline comparisons require compatible units, methods, habitat and temporal context
constraint C4: absence is not established by non-detection without accounting for detection probability
constraint C5: an association between indicator and pressure does not by itself establish causation
constraint C6: a taxon's indicator meaning is context-dependent, not universal across ecosystems
constraint C7: an ecological index must identify its component metrics and calculation method
constraint C8: missing data are distinct from measured zero values

# 5. INFERENCE RULES
rule R1 IndicatorResponse:
  IF observation(indicator, t, site) AND calibrated_response(indicator, pressure, context)
  THEN evidence_for(pressure, site, t) WITH confidence calibrated_from(response, uncertainty, context)

rule R2 BaselineDeviation:
  IF comparable(observation, baseline) AND deviation_exceeds_threshold(observation, baseline)
  THEN flag(environmental_change) WITH provenance(observation, baseline)

rule R3 MultiIndicatorConvergence:
  IF independent_indicators_support(condition) AND methods_compatible
  THEN strengthen_evidence(condition) WITHOUT assuming independence unless justified

rule R4 ConfounderCheck:
  IF alternative_driver_explains(response)
  THEN downgrade_specificity_of_inference AND record_confounder

rule R5 TrendInference:
  IF repeated_comparable_observations AND trend_uncertainty_is_estimated
  THEN classify_trend(improving | stable | deteriorating | indeterminate)

# 6. QUANTITATIVE MODELS
model StandardisedDeviation:
  z = (observed_value - baseline_mean) / baseline_standard_deviation
  requires: baseline_standard_deviation > 0 and comparable measurements

model RelativeChange:
  delta_relative = (value_t2 - value_t1) / abs(value_t1)
  requires: value_t1 != 0 and comparable time points

model OccupancyDetection:
  observed_detection_probability = occupancy_probability * detection_probability
  note: simplified single-visit expression; repeated surveys require explicit detection model

model WeightedCompositeIndex:
  index = sum_i(weight_i * normalised_metric_i) / sum_i(weight_i)
  requires: nonnegative weights, positive total weight, documented normalisation and direction

model BayesianEvidenceUpdate:
  posterior_odds = prior_odds * likelihood_ratio
  requires: calibrated likelihoods and explicit model assumptions

# 7. INVARIANTS
invariant I1: preserve observation identity and source provenance
invariant I2: preserve units and spatial-temporal scope through transformations
invariant I3: do not infer a pressure from a biological response without contextual evidence
invariant I4: all quantitative transformations expose assumptions and uncertainty
invariant I5: indicator assessments remain revisable when new observations arrive

# 8. CONFIDENCE
CONFIDENCE:
  inputs: [sampling_quality, detection_reliability, calibration_strength, spatial_representativeness, temporal_representativeness, confounder_control]
  output_range: [0, 1]
  aggregation: calibrated_model_required
  policy: never fabricate numerical confidence where calibration is unavailable

# 9. SEMANTIC EQUIVALENCE
SEMANTIC_EQUIVALENCE:
  compare: [indicator_kind, taxon_or_metric, habitat, pressure, method, units, spatial_scale, temporal_scale]
  exact: same operational definition and compatible scope
  related: overlapping concepts but non-interchangeable methods or contexts
  distinct: differing ecological meanings
  preserve_aliases: true

# 10. FLOW HASH
FLOW_HASH:
  canonicalise: [ontology_id, version, entity_types, relations, constraints, inference_rules, model_definitions]
  algorithm: SHA-256
  exclude: [display_formatting, runtime_timestamps]
  note: calculate only after deterministic serialisation; no digest is asserted here

# 11. UPDATE MODEL
UPDATE_MODEL:
  event_types: [new_observation, revised_baseline, taxonomic_revision, calibration_update, method_change, retracted_evidence]
  process:
    - validate event and provenance
    - retain previous version
    - identify dependent assessments
    - recompute affected inferences and uncertainty
    - record version, reason and timestamp
    - publish change log

# 12. AI CONSUMPTION
AI_CONSUMPTION:
  output_fields: [entity_id, type, indicator, target_condition, evidence, location, time, method, units, uncertainty, confidence, provenance, version]
  rules:
    - distinguish observation from inference
    - distinguish association from causation
    - label missing or unverified evidence
    - preserve source attribution
    - expose model limitations
    - do not silently merge incompatible baselines

# 13. VALIDATION CASES
case V1: an observed lichen decline with an appropriate baseline may support an air-quality assessment, but is not proof of one pollutant
case V2: macroinvertebrate diversity observations may contribute to freshwater status only with a specified index and sampling method
case V3: no species detected in one survey does not prove local extinction
case V4: two indicator values measured in incompatible units must not be directly compared
case V5: changing a baseline triggers reassessment of dependent classifications

# 14. CROSS-ONTOLOGY LINKS
links:
  - Biological Environment Ontology
  - Ecological Entity Ontology
  - Indicator Species Ontology
  - Habitat Ontology
  - Aquatic Habitat Ontology
  - Soil Habitat Ontology
  - Population Ontology
  - Biological Community Ontology
  - Environmental Measurement Ontology

# 15. LIMITATIONS
This document specifies a conceptual, typed Flow Algebra. It is not a deployed inference engine, a validated scientific indicator protocol, or a claim that the source webpage contains this exact syntax. Operational use requires data schemas, calibration, uncertainty estimation, executable semantics and scientific validation.

END FLOW