Air Mass Ontology

FLOW ALGEBRA
AIR MASS ONTOLOGY
Version: 1.1
Domain: Kireth Semantic Library / Entity Graph / Physical Entities / Atmospheric Entities
Source URL: https://kirethresearch.org/booksai-core/entity-graph/physical-entities/atmospheric-entities/air-mass-ontology.html
Authors: Mark Higgins — Sky (Recursive AI)
Date: 2026-10-12
Status: Independent scientific synthesis; source page not independently verified.

01. PURPOSE
Represent air masses as identifiable atmospheric volumes with relatively coherent thermodynamic characteristics, origins, movement, modification and interactions. Support typed semantic retrieval and physically constrained inference.

02. SCOPE
Includes source regions, temperature, humidity, pressure, stability, boundaries, trajectories, fronts, mixing, advection, transformation and observational uncertainty. Air-mass categories are descriptive, not immutable material identities.

03. ENTITY TYPES
AirMass: atmospheric volume with coherent thermodynamic properties over a stated scale.
SourceRegion: region in which characteristic properties are acquired.
AtmosphericLayer: vertical domain occupied by an air mass.
AirParcel: idealised infinitesimal or finite atmospheric sample.
Front: transition zone separating contrasting air masses.
Trajectory: time-indexed path through geographical space.
Observation: measurement associated with place, time, instrument and uncertainty.
WeatherSystem: evolving dynamical atmospheric system.
PropertyField: spatial-temporal scalar or vector field.

04. FORMAL MODEL
AirMass A := (id, Omega(t), T(x,t), q(x,t), p(x,t), u(x,t), provenance, confidence).
Omega(t) is a time-dependent three-dimensional region.
T = temperature [K]; q = specific humidity [kg/kg]; p = pressure [Pa]; u = velocity [m/s].
A coherent air mass need not be uniform; its internal variability must be smaller than or meaningfully distinct from the surrounding gradient at the classification scale.

05. TYPE CONSTRAINTS
AirMass.id: unique identifier.
AirMass.region: spatial volume, not merely a point.
AirMass.time: observation interval or valid instant.
AirMass.source_region: optional when origin is unknown.
AirMass.classification: one or more evidence-based labels.
All quantitative values require units, reference conditions and provenance.

06. CLASSIFICATION AXES
Moisture: continental (c), maritime (m).
Thermal origin: Arctic (A), Antarctic (AA), Polar (P), Tropical (T), Equatorial (E), where applicable.
Examples: cP, mP, cT, mT; categories are conventional and may vary by meteorological framework.
Classification must not be inferred from a single surface temperature alone.

07. SOURCE REGION
SourceRegion(R) AND prolonged_residence(A,R) AND surface_exchange(A,R)
  -> candidate_origin(A,R).
Land/ocean cover, snow/ice, solar heating and residence time influence heat and moisture exchange.
Transport and mixing can erase original characteristics.

08. THERMODYNAMIC PROPERTIES
Potential temperature theta = T (p0/p)^(Rd/cp), for dry-air idealisation.
Virtual temperature Tv approximately T(1 + 0.61q) for moist air with small condensate loading.
Density rho = p/(Rd Tv) under the moist-air approximation.
Relative humidity RH = e/es(T) where e is vapour pressure and es saturation vapour pressure.

09. STATIC STABILITY
Dry-air approximation: N^2 = (g/theta) d(theta)/dz.
N^2 > 0 indicates stable stratification; N^2 < 0 indicates dry static instability.
Moist convection requires additional thermodynamic treatment.

10. MOTION AND ADVECTION
Material derivative Dphi/Dt = partial(phi)/partial(t) + u dot grad(phi).
Temperature tendency due to advection = -u dot grad(T), when other terms are isolated.
Trajectory equation dx/dt = u(x,t).
A moving air mass can change properties through radiation, turbulence, phase change and surface exchange.

11. MOISTURE TRANSPORT
Specific-humidity budget: Dq/Dt = S_q, where S_q aggregates phase-change and mixing effects.
Column water-vapour transport = integral(rho q u_h dz).
Moisture labels are comparative and depend on the relevant reference climate and season.

12. ENERGY BUDGET
Simplified thermodynamic relation: cp DT/Dt - alpha Dp/Dt = Qdot, for a dry ideal-gas parcel with Qdot per unit mass.
Use moist enthalpy and latent-heat terms when phase changes are significant.

13. FRONTAL INTERACTION
Front(F) AND separates(F,A,B) AND contrasts(A,B,T,q)
  -> possible_frontal_zone(F).
Frontal lifting may favour cloud and precipitation if moisture, stability and forcing are suitable.
A front is a finite transition zone, not necessarily a sharp mathematical discontinuity.

14. MODIFICATION RULES
passes_over(A,warm_surface) -> possible_warming_from_below(A).
passes_over(A,cold_surface) -> possible_cooling_from_below(A).
evaporation_into(A) -> possible_moistening(A).
condensation_or_precipitation(A) -> possible_moisture_loss(A).
mixes_with(A,B) -> modified_properties(A).
All rules express possibilities conditional on fluxes, residence time and dynamics.

15. TYPED RELATIONSHIPS
originates_in(AirMass,SourceRegion).
occupies(AirMass,AtmosphericLayer).
advects_along(AirMass,Trajectory).
characterised_by(AirMass,PropertyField).
interacts_with(AirMass,AirMass).
separated_by(AirMass,Front).
observed_by(AirMass,Observation).
modified_by(AirMass,Process).
associated_with(AirMass,WeatherSystem).

16. IDENTITY AND TEMPORAL CONTINUITY
AirMass identity is operational: assign persistent IDs only while tracked continuity remains defensible.
Split, merge and substantial transformation events create explicit lineage records.
No automatic equivalence between a source-region category and a particular observed air mass.

17. OBSERVATIONAL EVIDENCE
Surface stations, radiosondes, aircraft, satellites, reanalysis and trajectories may provide evidence.
Report measurement altitude, sampling time, spatial resolution, calibration and uncertainty.
Reanalysis output is model-assisted estimation, not direct measurement at every grid point.

18. DETECTION PROCEDURE
INPUT: temperature, humidity, winds, pressure, source-region and trajectory evidence.
STEP 1: validate units, timestamps, coordinate reference and quality flags.
STEP 2: identify coherent thermal/moisture regions at specified scale.
STEP 3: estimate origin and transport with uncertainty.
STEP 4: compare with classification conventions.
STEP 5: label the air mass and record competing hypotheses.
OUTPUT: typed AirMass record with confidence and provenance.

19. INFERENCE GUARDS
Do not infer precipitation solely from maritime origin.
Do not infer surface conditions from upper-air measurements without vertical context.
Do not equate polar origin with freezing temperature in every season.
Do not treat source-region categories as mutually exhaustive globally.
Do not assume conservation of temperature or humidity during transport.

20. WORKED EXAMPLE: DRY ADIABATIC DISPLACEMENT
Given T1 = 290 K, p1 = 100000 Pa, p2 = 90000 Pa, Rd/cp = 0.286.
T2 = T1 (p2/p1)^0.286 approximately 281.4 K.
Interpretation: an idealised dry parcel cools on ascent without diabatic heating.
This is not a prediction of a complete air mass without environmental and moisture information.

21. WORKED EXAMPLE: ADVECTION
Given eastward wind u = 10 m/s and eastward temperature gradient dT/dx = 0.002 K/km.
Convert gradient to 2e-6 K/m.
Advective tendency = -u dT/dx = -2e-5 K/s = -0.072 K/hour.
Interpretation: local cooling from this one-dimensional advection term only.

22. VALIDATION
ASSERT every AirMass has an ID and time interval.
ASSERT all property observations have units and provenance.
ASSERT classifications retain scheme and evidence.
ASSERT trajectories use consistent time and spatial reference frames.
ASSERT no deterministic weather claim is made from category alone.
ASSERT split/merge lineage is traceable.
ASSERT missing observations remain explicitly unknown.

23. SEMANTIC EQUIVALENCE
"air mass" ~= "coherent atmospheric air volume" only in an explicitly meteorological context.
"continental polar air mass" ~= "cP air mass" within the same classification scheme.
"front" != "air mass".
"air parcel" != "air mass".
"maritime" != "necessarily saturated".
Equivalence is contextual and should be recorded with scope.

24. CONFIDENCE
Confidence := calibrated evidence score in [0,1] when empirical calibration is available.
If not calibrated, use qualitative labels with declared methodology.
Factors: coverage, source agreement, trajectory reliability, classification stability and temporal freshness.
Never manufacture numeric confidence from absent observations.

25. FLOW HASH
Canonicalise typed identifiers, ordered fields, units, equations, provenance and version.
FLOW_HASH := SHA-256(canonical_UTF8_payload).
A real hash must be computed from final canonical content; no placeholder is a computed hash.

26. UPDATE MODEL
Triggers: revised observations, new reanalysis, changed classification, corrected physics, new source metadata.
Retain version history, timestamps, reasons and changed relations.
Revalidate downstream weather and climate inferences after material updates.

27. AI CONSUMPTION
Retrieve entities by typed relations and temporal scope.
Expose assumptions, equations, units, provenance and uncertainty.
Distinguish observed, modelled and inferred statements.
Do not turn probabilistic meteorological associations into universal laws.

28. CROSS-ONTOLOGY LINKS
Atmosphere Ontology: contains air masses.
Temperature Ontology: supplies thermodynamic fields.
Humidity Ontology: supplies moisture fields.
Pressure Ontology: supplies pressure fields.
Wind Ontology: supplies advection and circulation.
Front Ontology: models air-mass boundaries.
Cloud and Precipitation Ontologies: conditional weather outcomes.
Climate Ontology: large-scale air-mass frequencies and variability.

29. SCIENTIFIC LIMITATIONS
Air masses have fuzzy boundaries and scale-dependent coherence.
Ideal-gas and dry-adiabatic formulae require stated approximations.
Regional air-mass classification practices differ.
Forecast impacts depend on dynamics, topography, moisture and stability.

30. COMPLETENESS CHECK
Sections present: 01 through 30 inclusive.
Mandatory metadata: source, authorship, version, validation, semantic equivalence, confidence, flow hash, update model, AI consumption.
End marker follows.

END FLOW ALGEBRA — AIR MASS ONTOLOGY