This layer maps agricultural systems as living relational systems under the HIR × OAM framework. HIR governs evidence boundaries, claim discipline, and life-first alignment. OAM maps degradation pressure, extraction pathways, feedback failure, repair friction, and cascade risk across soil, water, biology, food systems, and ecological boundaries.
Agriculture is the planetary-life continuation of the abiogenesis / Big Bloom cycle: soil and water as the chemical substrate, microbes and fungi as the metabolic network, plants and animals as the organized biological production layer, humans as participants and stewards, and waste, death, compost, and decomposition as the return pathway that enables renewal. A food system that breaks any of these loops is consuming its own future.
| In scope | Out of scope |
|---|---|
| Degradation pathway mapping for agricultural systems | Site-specific farming prescriptions of any kind |
| OAM variable mapping to soil, water, biology, food systems | Crop treatment, pesticide, fertilizer, or irrigation guidance |
| HIR evidence boundaries per domain and measurement | Financial advice or economic recommendation |
| Measurement/provenance uncertainty per agricultural signal | Replacement for agronomy, soil science, ecology, or farming expertise |
| Degradation cascade analysis (Ξ across soil/water/ecology) | Deterministic predictions of farm outcomes |
| Life-first regeneration framework (vs. yield-only framing) | Claims of "regeneration" without time-series evidence |
| Non-moralizing systems-pressure analysis | Farmer blame, shame, or character claims of any kind |
| Variable / Gate | Agricultural meaning | What it constrains or maps |
|---|---|---|
| HIR — H (Honesty) | Measured condition vs. assumed condition. Provenance of soil, water, yield, climate, labor, and supply-chain data. | Label what is directly measured vs. inferred. Single measurement ≠ trend. Remote sensing ≠ ground truth. Farmer observation must be provenance-labeled. Unknown mechanism stays unknown. |
| HIR — I (Integrity) | Whether soil, water, nutrients, microbes, biodiversity, animals, humans, and production remain structurally aligned over time. | Yield ≠ soil health. Productivity ≠ sustainability. Proxy ≠ system truth. Fertilizer response ≠ health. Pesticide success ≠ ecological stability. Monoculture efficiency ≠ resilience. |
| HIR — R (Respect) | Boundary-respecting relation among soil life, water cycles, pollinators, animals, farmers, workers, consumers, future generations, and local ecology. | No farmer blame or moralization. Systems pressure must be separated from individual blame. No prescription without local data. No intervention claim without evidence. |
| HIR — Fidelity | Honest signal + structural consistency: measurement provenance + system alignment over time. | A fidelity signal in agriculture: soil carbon measured by consistent method over multiple years, with known sampling depth and seasonal context. |
| HIR — Cohesion | Mutually supporting relation between soil, water, biology, climate, labor, economics, and food access. | A cohesive agricultural system: water retention supports soil biology supports crop roots supports watershed health supports community food security — each layer reinforces, not cannibalizes. |
| HIR — Resonance | Life-compatible agricultural function that endures pressure while preserving repair capacity. | Resonant agriculture is not maximum-extraction. It is production that does not destroy the conditions required for future life. Rn > 0 sustained over agricultural time scales. |
| OAM — Degradation Engine · equations.py | ||
| OAM — P (Pressure) | W = acute stressor (drought event, pest outbreak, flood, market shock). F = chronic load (soil depletion rate, water overdraft, monoculture fragility, sustained chemical dependency, labor burden). WF = multiplicative coupling. | P = w_W·W + w_F·F + w_WF·WF. A farm under chronic depletion (F high) that also experiences acute drought (W high) faces P > sum of either alone. |
| OAM — D (Damage) | Cumulative agricultural degradation: soil carbon loss, aquifer drawdown, microbial disruption, biodiversity loss, erosion, salinization, topsoil depth decline, pollinator loss, food-system fragility. | D grows at growth_per_cycle when P > repair. D is floored at 0. D is invisible in yield data during production masking phase. |
| OAM — C (Reserve) | Biological reserve: soil organic matter depth, aquifer level, seed genetic diversity, microbial biomass, mycorrhizal network extent, pollinator population, buffer land, ecological margin. | C depletes when D accumulates faster than repair. When C approaches 0, system cannot sustain production without escalating external inputs. |
| OAM — Θ (Repair) | Biological repair traction: microbial/fungal activity rate, organic matter decomposition and incorporation, water cycle restoration, biodiversity recovery, cover-crop ground retention, restoration lag. | Θ = σ(−1 + θ_C·C + θ_E·E − θ_K·K). Repair traction rises with C (reserve) and E (compost, rain, biodiversity access); falls with K (compaction, salinity, pesticide toxicity, economic pressure). |
| OAM — K (Resistance) | Resistance to repair: soil compaction (mechanical barrier to root and microbe), salinization (toxic to soil biology), pesticide persistence (disrupts repair biology), market pressure locking in extractive practices, debt-driven input dependency. | K rises as D accumulates. High-K soils cannot repair even when E (resources) increases, because physical/chemical/economic barriers block Θ activation. |
| OAM — Ξ (Cascade) | Degradation propagating across agricultural domains: soil compaction → water infiltration failure → drought stress → crop loss → chemical input escalation → soil biology disruption → more compaction (vicious loop). Eutrophication: excess nutrient runoff → aquatic hypoxia → fishery loss → community food loss. | Ξ = Ξ_base + (σ·Ξ_unit·Act)·Λ. Cascade propagation is why single-domain intervention is insufficient in advanced degradation. |
| OAM Failure Signal | Production remains visible while repair capacity declines invisibly. Yield maintained by external inputs while C → 0. Aquifer drained while crops grow normally. Topsoil lost while satellite data shows green fields. | This is the most dangerous agricultural failure mode: it appears successful until threshold crossing. D has been accumulating for years before visible collapse. |
| HIR Gate | Blocks overclaiming: yield ≠ health, productivity ≠ sustainability, proxy ≠ system truth, single measurement ≠ trend, remote sensing ≠ ground truth, prescription without local data = blocked. | intervention_claim_allowed = false by default. site_specific_claim_allowed = false without local data. hard_override_triggered = true if prescription appears in output without local data. |
| ID | Pathway | Affected domains | OAM degradation | HIR boundary |
|---|---|---|---|---|
| AG-P-01 | Soil Erosion | AG-D-01 AG-D-03 | P = wind/water/tillage exposure. D = topsoil depth loss, organic matter loss. K = bare soil, crusting. Ξ → AG-D-03 (runoff), AG-D-05 (root access loss) | Erosion rate estimate ≠ projection. LOCAL_DATA_REQUIRED for rate quantification. Erosion ≠ simple reversal. |
| AG-P-02 | Soil Compaction | AG-D-01 AG-D-03 AG-D-02 | P = heavy machinery, overgrazing, tillage on wet soil. D = pore collapse, bulk density increase. K = hardpan formation (very high K — resists most repair). Ξ → infiltration failure → drought vulnerability | Compaction depth and extent requires on-site penetrometer data. Cannot be inferred from yield or satellite alone. |
| AG-P-03 | Organic Matter Depletion | AG-D-01 AG-D-02 AG-D-04 | P = tillage, oxidation, residue removal. F = continuous bare-soil cropping. D = soil carbon, microbial biomass, aggregate stability. C declines (C = SOM reserve). Ξ → nutrient availability, water retention, erosion | SOM increase ≠ confirmed regeneration without time series. Soil carbon proxy ≠ complete health. AG-R-04 applies. |
| AG-P-04 | Microbial / Fungal Disruption | AG-D-02 AG-D-04 | P = pesticide, tillage, compaction, salinity. D = microbial biomass, mycorrhizal network extent, functional diversity. K = persistent chemical residues. Ξ → nutrient cycling failure, plant stress | Microbial function is highly complex (AG-UA-01). Single assay ≠ soil food web health. AG-UA-02 applies throughout. |
| AG-P-05 | Nutrient Runoff / Eutrophication | AG-D-04 AG-D-03 AG-D-16 | P = fertilizer over-application, bare soil, impervious surface. F = chronic excess nutrient loading to water bodies. Ξ = algal bloom → hypoxia → aquatic life loss → fishery collapse → food access impact | Eutrophication attribution requires watershed-level data. Single-field measurement insufficient for downstream impact claim. |
| AG-P-06 | Salinization | AG-D-01 AG-D-03 | P = irrigation with high-salt water, capillary rise in arid climates, evaporation concentration. D = soil EC increase, clay dispersion. K = established saline profile (very high K — decades for natural leaching). Ξ → crop loss, land abandonment | Salinization severity requires EC measurement by depth. Remediation timelines LOCAL_DATA_REQUIRED. No prescription claim. |
| AG-P-07 | Groundwater Depletion | AG-D-03 | P = irrigation withdrawal rate exceeding recharge. D = water-table drawdown, aquifer storage loss. K = geological limits on recharge. No biological repair mechanism for fossil aquifers. Ξ → land subsidence, loss of future agricultural capacity | Aquifer depletion rate requires hydrological data LOCAL_DATA_REQUIRED. Water table depth is one signal; recharge rate requires separate measurement. |
| AG-P-08 | Drought Stress | AG-D-03 AG-D-05 AG-D-02 | P = W (acute) or F (chronic). Drought stress on already-degraded soils (low C, compacted) is amplified vs. healthy soils. Ξ = drought → soil biology collapse → reduced infiltration → worse drought vulnerability | Drought impact depends on soil health baseline. Crop stress is not single-cause (AG-R-11). LOCAL_DATA_REQUIRED for severity assessment. |
| AG-P-09 | Heat Stress | AG-D-05 AG-D-07 AG-D-12 | P = W (acute heat events), F (rising mean temperature). D = photosynthetic efficiency loss, pollinator phenology mismatch, labor capacity reduction, food safety risk. C = local cooling from tree cover, healthy soil water retention | Heat stress attribution requires site temperature records. AG-UA-03 (climate interaction) applies throughout. |
| AG-P-10 | Monoculture Fragility | AG-D-05 AG-D-06 AG-D-09 AG-D-13 | P = F (chronic genetic and ecological narrowing). K = high genetic uniformity (a single pathogen can propagate across entire production region). Ξ = disease outbreak → regional crop loss → supply-chain collapse. C = genetic diversity reserve ↓ | Monoculture efficiency ≠ resilience (AG-R-09). AG-UA-06 (biodiversity function uncertain) applies. Resistance quantification requires pathological testing. |
| AG-P-11 | Pesticide Resistance | AG-D-09 AG-D-10 | P = selection pressure from repeated chemical application. D = resistant pest/pathogen population prevalence. K = resistance genes (cannot be reversed by continued same-chemistry use). Ξ = resistance → escalating dose → biology disruption → secondary pest outbreaks | Pesticide success ≠ ecological stability (AG-R-08). Resistance status requires local surveillance data. No pesticide recommendation. |
| AG-P-12 | Herbicide Dependency | AG-D-09 AG-D-10 AG-D-02 | P = F (chronic herbicide use). D = soil biology disruption, weed community shift toward tolerant species. K = herbicide-tolerant weed populations. Ξ → soil health further degraded, microbiome altered | No herbicide recommendation or characterization. Herbicide response ≠ proof of system health (AG-R-07). LOCAL_DATA_REQUIRED. |
| AG-P-13 | Pollinator Decline | AG-D-07 AG-D-05 AG-D-16 | P = pesticide exposure, habitat loss, pathogen load, monoculture (low floral diversity). D = pollinator population and diversity. C = wild pollinator reserve, habitat connectivity. Ξ → crop pollination failure → yield decline → food access impact | Pollinator decline is multifactorial (AG-UA-04). Single pesticide class ≠ sole cause. Local pollinator surveys needed. |
| AG-P-14 | Biodiversity Collapse | AG-D-16 AG-D-07 AG-D-02 AG-D-09 | P = habitat simplification, pesticide, monoculture, invasive species. D = species richness, functional diversity. K = habitat fragmentation (patches too small for population viability). Ξ → cascade ecosystem service losses | Biodiversity loss may remain hidden until threshold effects appear (AG-R-12). AG-UA-09 (ecosystem threshold unknown) applies. AG-UA-06 applies. |
| AG-P-15 | Livestock Waste Concentration | AG-D-08 AG-D-03 AG-D-04 | P = confinement density beyond land processing capacity. D = water contamination (nitrate, pathogens, pharmaceuticals), soil overload, methane, ammonia. K = regulatory/economic constraints on dispersion. Ξ → watershed contamination, food safety risk | No livestock management prescription. Waste concentration risk requires watershed and density data LOCAL_DATA_REQUIRED. |
| AG-P-16 | Overgrazing | AG-D-08 AG-D-01 AG-D-02 AG-D-16 | P = stocking rate exceeding pasture recovery rate. D = root system loss, bare soil exposure, erosion, compaction, microbial disruption. K = established bare-soil state. Ξ → erosion → water quality → riparian damage | Overgrazing threshold is site-specific (forage, rainfall, season). LOCAL_DATA_REQUIRED. No stocking rate recommendation. |
| AG-P-17 | Crop Disease Amplification | AG-D-05 AG-D-09 AG-D-10 | P = genetic uniformity + stressed plant (low U) + pathogen inoculum. D = crop loss, soil inoculum accumulation. K = established soil pathogen load. Ξ = disease → increased fungicide → soil biology disruption → plant stress → more disease | Crop disease is multifactorial (AG-UA-04). No fungicide recommendation. Disease severity requires lab identification LOCAL_DATA_REQUIRED. |
| AG-P-18 | Supply-Chain Fragility | AG-D-13 AG-D-14 | P = concentration of processing, transport, and retail; geographic/commodity concentration; input dependency. D = system fragility (few nodes, high impact if one fails). K = consolidation prevents easy redistribution. Ξ = supply shock → food price spike → food access loss | Supply-chain efficiency ≠ food-system resilience (AG-R-15). Fragility assessment requires systems-level data AG-UB-09. |
| AG-P-19 | Food Insecurity | AG-D-14 AG-D-13 AG-D-12 | P = production instability, supply-chain failure, affordability barriers, distribution inequality. D = dietary diversity, nutritional health of populations. K = structural poverty, lack of food sovereignty, market dependency. Ξ = food insecurity → health burden → labor capacity loss → productivity decline | Food insecurity is systemic and political. No individual-level dietary recommendation. No policy recommendation without local economic and equity context AG-UB-10. |
| AG-P-20 | Yield-over-Health Optimization | AG-D-01AG-D-02AG-D-03AG-D-04AG-D-16 | See expanded pathway below. This is the systemic optimization failure: all other degradation pathways are in part driven by maximizing yield as the primary metric without accounting for the repair/substrate-return costs. | Yield is not health (AG-R-01). Productivity is not sustainability (AG-R-02). Yield increase ≠ system improvement without soil/water/biodiversity data. |
| AG-P-21 | Ecological Debt Accumulation | AG-D-16 AG-D-04 AG-D-03 | P = externalization of biological costs (ecosystem services consumed without replacement). D = accumulated ecological services deficit: pollination, water filtration, carbon sequestration, flood buffering lost. K = irreversible losses (extinct species, fossil aquifer depletion). This is OAM D at planetary scale. | Ecological debt cannot be measured by single farm metrics. Requires ecosystem-level monitoring AG-UA-07. No "offset" claim without rigorous measurement and additionality assessment. |
| ID | Measurement | Directly measures | May suggest | Cannot prove alone | Quality risks | Confidence impact |
|---|---|---|---|---|---|---|
| AG-M-01 | Soil organic matter (%) | % organic material by weight/volume at sampling depth and date | Biological activity, water retention, structural stability trend | Soil health in full; microbial function; long-term trend (single point) | Sampling depth, season, method, lab consistency all affect comparability | AG-UB-02 AG-UB-03 |
| AG-M-02 | Soil carbon (g/kg, %) | Total or organic carbon concentration at sampling depth | SOM trend, carbon sequestration potential | Complete soil health; climate benefit without additionality measurement; trend without time series | Method (LOI vs. combustion), depth, bulk density correction required for stock calculations | AG-UB-03 |
| AG-M-03 | Soil pH | Hydrogen ion concentration in soil solution at measurement time | Nutrient availability, biological activity range | Nutrient availability without nutrient testing; liming need without soil type + crop context | Seasonal variation, water content, lab method | AG-UB-02 |
| AG-M-04 | N/P/K (nitrogen, phosphorus, potassium) | Plant-available N, P, K at measurement time | Fertilization response potential; nutrient cycling status | Long-term fertility trend (single test); optimal application rate without yield goal and local context | Temporal variation high (especially N); seasonal confounds; extraction method varies | AG-UB-01 AG-UB-05 AG-UB-12 |
| AG-M-05 | Micronutrients (Fe, Zn, Cu, Mn, B, Mo) | Plant-available micronutrient concentrations at sampling depth | Deficiency risk; crop-specific availability | Deficiency confirmation without plant tissue test; application need without crop × soil type context | Extraction method, pH interaction, crop-specific availability varies | AG-UB-01 |
| AG-M-06 | Soil moisture (%) | Volumetric or gravimetric water content at depth and time | Plant-available water, infiltration status, irrigation timing signal | Drought stress without crop context and phenological stage; irrigation need without ET data | Point measurement only; high spatial variability; sensor calibration required | AG-UB-04 AG-UB-05 |
| AG-M-07 | Infiltration rate (mm/hr) | Rate of water entry into soil at measurement point | Compaction status, organic matter trend, runoff potential | Full soil structure; long-term trend (single measurement); compaction depth without penetrometer | High spatial variability; method (ring/tension infiltrometer) affects results | AG-UB-03 |
| AG-M-08 | Compaction (penetration resistance, MPa) | Mechanical resistance to probe penetration at depth | Tillage pan location, root penetration barrier depth | Compaction cause; remediation need without biological, water, and crop context; single probe reading represents only that point | Moisture-dependent; soil type interaction; single probe is a point measurement | AG-UB-03 AG-UB-05 |
| AG-M-09 | Erosion rate (t/ha/yr estimate) | Estimated mass of soil lost per unit area per year | Topsoil depth decline risk, water quality risk downstream | Actual soil lost without validation measurement; downstream impact without watershed data | USLE/RUSLE model estimates widely used but model assumptions may not match all conditions LOCAL_DATA_REQUIRED | AG-UB-03 AG-UB-05 |
| AG-M-10 | Microbial activity (respiration, PLFA, enzyme assays) | Biological activity rate or microbial community composition proxy at sampling | Soil biological health trend, organic matter processing rate | Full soil food web function (AG-UA-02); specific organisms or processes; trend without time series | Highly sensitive to sampling, handling, temperature, moisture at time of collection | AG-UA-01 AG-UA-02 AG-UB-03 |
| AG-M-11 | Fungal/mycorrhizal indicators (hyphal length, spore counts) | Presence and density of fungal hyphae or mycorrhizal spores in sample | Mycorrhizal colonization potential, network connectivity | Functional connectivity; crop benefit without colonization assessment and plant tissue analysis; AG-UA-01 | Very sensitive to disturbance; high spatial variability; method-dependent | AG-UA-01 AG-UA-02 |
| AG-M-12 | Water-table depth (m bgl) | Depth to saturated zone at measurement well/time | Aquifer depletion trend (if time series), irrigation capacity | Recharge rate without hydrological study; depletion rate without historical data; regional trend from single well | Single well ≠ aquifer; seasonal variation; well interference from neighboring pumping | AG-UB-11 |
| AG-M-13 | Irrigation volume (m³ or ML applied) | Volume of water applied per season or per unit area | Water use intensity, overdraft risk (if compared to ET + precipitation) | Efficiency without soil moisture, ET, and crop uptake data; sustainability without recharge data | Meter accuracy; missing records common AG-UB-06 | AG-UB-06 AG-UB-11 |
| AG-M-14 | Salinity (EC, dS/m) | Electrical conductivity of soil solution or irrigation water | Salt accumulation risk, crop tolerance threshold comparison | Origin of salinity without source testing; remediation need without full profile and drainage context | Depth profile required; seasonal variation; leaching fraction calculation needed | AG-UB-11 |
| AG-M-15 | Crop yield (t/ha, bu/ac) | Harvested mass per unit area in one season | Production level at time of measurement | Soil health (AG-R-01). Sustainability (AG-R-02). Future production without soil and water trend data. | Variety, season, inputs, and weather all confound year-to-year comparisons without control | AG-UA-05 AG-UB-05 |
| AG-M-16 | Crop nutrient density | Nutritional content of harvested product (protein, minerals, vitamins) | Food quality trend if time series with consistent variety and method | Cause of change without soil, variety, and management data; public health impact without diet context | Highly method- and variety-dependent; post-harvest changes; limited longitudinal data REVIEW_REQUIRED | AG-UA-05 |
| AG-M-17 | Pesticide / herbicide records | Application products, rates, timing as recorded | Chemical load on field, resistance selection pressure | Residue levels without soil/water/tissue testing; ecological impact without monitoring | Records often incomplete AG-UB-06; does not include drift or off-target effects | AG-UB-06 |
| AG-M-18 | Biodiversity counts (species richness, abundance) | Number and abundance of identified species in a survey area at survey time | Diversity trend if time series; functional group representation | Ecosystem function (AG-UA-06); full species complement (detection limit); trajectory without repeat surveys | Highly survey-method dependent; seasonal and weather effects; observer skill | AG-UA-06 AG-UB-03 |
| AG-M-19 | Pollinator counts | Observed pollinator visits or captures per unit area per time | Pollinator abundance trend; species richness at site | Crop pollination adequacy without actual pollination observation; population trend without multi-year data | Time of day, weather, floral availability, season all affect counts | AG-UA-04 |
| AG-M-20 | Satellite / remote sensing | Spectral reflectance data at sensor resolution and overpass timing | NDVI trend, crop stress indicators, land cover change | Soil health below canopy; moisture status without calibration; yield without ground-truthing | Resolution limits; cloud cover; atmospheric correction; calibration required for cross-date comparison | AG-UB-07 |
| AG-M-21 | Weather data | Temperature, precipitation, humidity, wind at station location | Growing season conditions, drought or flood risk context | Farm-level conditions without on-site station; long-term climate trend from short record | Station distance from farm; microclimate variation; gaps in record | AG-UB-05 |
| AG-M-22 | Farmer observations | Experienced practitioner's direct observations of farm system over time | Local patterns, historical context, management response history | Objective measurement without provenance; comparative claim without calibration | Must be provenance-labeled (AG-R-06); valuable but subject to recall and framing bias | AG-UB-08 |
| AG-M-23 | Supply-chain data | Volumes, prices, origins, transport routes, processing steps in documented chain | Concentration risk, fragility, provenance of produce | Food safety without testing; environmental impact without LCA; labor conditions without direct audit | Often proprietary, incomplete, or aggregated; verification difficult AG-UB-09 | AG-UB-09 |
| AG-M-24 | Labor data | Employment levels, wage records, labor conditions as documented | Labor system health, stewardship capacity | Worker wellbeing without direct assessment; knowledge holder capacity without survey | Often incomplete; self-reported; labor conditions vary by crop and region | AG-UB-10 |
| AG-M-25 | Economic data | Revenue, input costs, debt levels, market prices at documented time | Financial pressure on farm system, input dependency | System health without soil/water/ecology data; sustainability without non-financial metrics | Financially viable farm ≠ ecologically resilient farm; lag between financial and ecological collapse | AG-UB-10 |
| AG-M-26 | Food access data | Affordability, availability, dietary diversity metrics in a defined population | Food security level, nutritional adequacy, equity of distribution | Cause of insecurity without economic and distribution analysis; health impact without dietary survey | Geographic and demographic aggregation may mask local disparities | AG-UB-10 |
| ID | Class | Cat. | Effect on interpretation | Downgrade? | Suspend? | Raise confidence? |
|---|---|---|---|---|---|---|
| AG-UA-01 | microbial_function_unknown | UA | Specific microbial species and their functional contributions to soil health are not fully characterized; biomass measurement ≠ function | Yes | No | Never |
| AG-UA-02 | soil_food_web_complexity | UA | Soil food web interactions (bacteria, fungi, protozoa, nematodes, arthropods) are highly complex and context-specific; simplified indicators are partial proxies only | Yes | No | Never |
| AG-UA-03 | climate_interaction_uncertain | UA | Local climate-soil-crop interactions are not fully predictable from regional climate data; individual farm response to climate event is not determined by average | Yes | No | Never |
| AG-UA-04 | pest_pressure_multifactorial | UA | Pest and pathogen outbreaks result from multiple interacting factors (host, pest, environment, management history); single-cause attribution is rarely valid | Yes | No | Never |
| AG-UA-05 | crop_response_context_dependent | UA | Crop response to any input, stress, or management change is highly context-dependent: variety, soil type, climate, prior history all modify response | Yes | No | Never |
| AG-UA-06 | biodiversity_function_uncertain | UA | The functional contribution of specific species or diversity levels to ecosystem services is not fully characterized; species richness is a partial proxy | Yes | No | Never |
| AG-UA-07 | long_term_regeneration_rate_unknown | UA | Rate of soil organic matter recovery, aquifer recharge, biodiversity restoration, and ecosystem function restoration is highly site-specific and often decades-long | Yes | Conditional | Never |
| AG-UA-08 | farm_specific_variability | UA | Farm-level outcomes cannot be predicted from regional averages; soil, topography, history, management, and market context all vary at the farm level | Yes | No | Never |
| AG-UA-09 | ecosystem_threshold_unknown | UA | Threshold at which gradual degradation produces sudden ecological collapse (tipping point) is not predictable from current data for most agricultural ecosystems | Yes | No | Never |
| AG-UA-10 | restoration_lag_unknown | UA | Time lag between management change and measurable biological response is highly variable; short-term metrics may not reflect long-term trajectory | Yes | Conditional | Never |
| CATEGORY AG-UB — MEASUREMENT / PROVENANCE | ||||||
| AG-UB-01 | missing_soil_test | UB | No baseline soil data available; cannot assess change, trend, or response without measurement | Yes | Yes | Never |
| AG-UB-02 | outdated_measurement | UB | Measurement is from a period that may not represent current conditions; soil, climate, and management change over time | Yes | Conditional | Never |
| AG-UB-03 | sampling_bias | UB | Sample location, depth, timing, or method may not represent the field or watershed being described; spatial and temporal variability often undersampled | Yes | Conditional | Never |
| AG-UB-04 | sensor_uncalibrated | UB | On-farm sensor data (moisture, temperature, conductivity) is unreliable without proper calibration to local soil type and conditions | Yes | Yes | Never |
| AG-UB-05 | weather_context_missing | UB | Soil and crop measurements without corresponding weather context (rainfall, temperature, extreme event) cannot be interpreted correctly | Yes | Conditional | Never |
| AG-UB-06 | incomplete_application_records | UB | Fertilizer, pesticide, irrigation, or tillage records are missing or incomplete; actual chemical load and management history unknown | Yes | Conditional | Never |
| AG-UB-07 | remote_sensing_resolution_limit | UB | Satellite or aerial imagery cannot resolve sub-field variability or below-canopy conditions; spectral proxies have site-specific calibration requirements | Yes | Conditional | Never |
| AG-UB-08 | farmer_observation_only | UB | Farmer observation is valuable but requires provenance labeling; cannot substitute for measured data in system health claims | Yes | No | Never |
| AG-UB-09 | supply_chain_data_unverified | UB | Supply-chain provenance data is self-reported or from third parties without independent verification; cannot support strong origin or environmental claims | Yes | Conditional | Never |
| AG-UB-10 | economic_context_missing | UB | Agricultural outcomes cannot be interpreted without economic context (debt, market price, subsidy, input cost); productivity without economics may hide systemic pressure | Yes | No | Never |
| AG-UB-11 | water_source_unknown | UB | Origin and sustainability of water source (surface vs. groundwater, seasonal vs. perennial, recharge rate) is not documented; water security cannot be assessed | Yes | Yes | Never |
| AG-UB-12 | land_history_unknown | UB | Prior land use, contamination history, prior crop types, and management practices are not documented; current measurements cannot be interpreted without baseline context | Yes | Conditional | Never |
// Agriculture Degradation / Regeneration Record — Primordial Agriculture Layer v0.1 { // --- Identity --- "record_id": "string", "source_id": "string // cited data source(s)", "domain_id": "enum[] // AG-D-01 through AG-D-16", "pathway_id": "enum // AG-P-01 through AG-P-21", "created_at": "ISO8601", "schema_version": "string", // --- Scope and Location --- "farm_or_region_scope": "enum // field | farm | watershed | region | supply_chain | unknown", "location_precision": "enum // exact | approximate | regional | withheld | unknown", "land_history_documented": "bool // false = AG-UB-12 active", // --- OAM Pathway Fields --- "pressure_sources": "string[] // W and F components described; W = acute, F = chronic", "system_strain": "string // how P manifests as soil/water/biology strain", "degradation_pathway": "string // mechanism by which D accumulates", "feedback_failure": "string // where Θ breaks or K rises; production masking pattern if present", "oam_P_estimate": "enum // low | moderate | high | very_high | unknown", "oam_D_estimate": "enum // negligible | mild | moderate | severe | unknown", "oam_K_estimate": "enum // low | moderate | high | unknown // compaction, salinity, toxicity, economic", "oam_Theta_estimate": "enum // high | moderate | low | absent | unknown // repair traction", "oam_C_estimate": "enum // preserved | reduced | significantly_reduced | unknown // SOM, aquifer, diversity", "production_masking_suspected": "bool // yield stable while C declining; D invisible in yield data", // --- Evidence --- "measurable_signs": [ { "measurement_id": "enum // AG-M-01 through AG-M-26", "value_or_description": "string // what was measured", "uncertainty_classes": "enum[] // AG-UB classes active for this measurement", "single_point_only": "bool // false = time series; true = AG-R-03 applies", "provenance_documented": "bool // false = AG-UB-08 applies if farmer obs" } ], "downstream_effects": "string[] // potential associated downstream effects — not predictions", // --- Repair / Regeneration --- "repair_pathways": "string[] // observed or described repair mechanisms — not prescriptions", "regeneration_status": "enum // not_assessed | early_signal | time_series_supported | REVIEW_REQUIRED", "substrate_return_documented": "bool // AG-D-15 loop: is organic matter return being measured", // --- Uncertainty --- "biological_unknown_class": "enum[] // AG-UA-01 through AG-UA-10", "measurement_unknown_class": "enum[] // AG-UB-01 through AG-UB-12", "evidence_level": "enum // none | anecdotal | single_measurement | multi_metric | time_series | expert_reviewed", "provenance_class": "enum // verified | partially_documented | unverified", "confidence_level": "enum // high | moderate | low | suspended", "interpretation_status": "enum // allowed_bounded | unresolved | contaminated | overclaim_blocked | REVIEW_REQUIRED", // --- Hard Gates (all default to the safe value) --- "intervention_claim_allowed": "bool // ALWAYS false — no prescription", "site_specific_claim_allowed": "bool // false unless local data + expert review confirmed", "regeneration_confirmed": "bool // false unless time_series evidence + expert review", "hard_override_triggered": "bool // true = prescription appeared or AG-R violated", // --- Governance Notes --- "HIR_boundary_note": "string // what HIR blocks for this record; prohibited inferences listed", "OAM_pathway_note": "string // OAM variable estimates and degradation structure", "notes": "string[]", "source_needed": "string[] // REVIEW_REQUIRED / SOURCE_NEEDED items", "linked_source_records": "string[] // citations and data provenance" }
site_specific_claim_allowed = false AND output contains prescription → hard_override_triggered = trueevidence_level < time_series AND regeneration_status = confirmed → interpretation_status = overclaim_blockedyield_increase = documented AND soil/water/biodiversity data missing → classification = unresolved / insufficient_data
| Stage | What happens | OAM variables | HIR governance |
|---|---|---|---|
| 1. Pressure Accumulation | Heat, drought, extraction, chemical dependency, market pressure, labor burden, water drawdown, and monoculture fragility accumulate over seasons. P = w_W·W (acute events) + w_F·F (chronic load) + coupling. | P rises. W = drought, pest, flood. F = ongoing depletion rate, chemical burden, economic pressure. | H: document each pressure source. I: distinguish acute from chronic. R: do not assign pressure to farmer character. |
| 2. Substrate Strain | Soil structure weakens. Organic matter declines. Microbial/fungal networks disrupted. Water retention decreases. Biodiversity narrows. Human stewardship capacity diminishes under economic pressure. | C (reserve) begins to decline: SOM, aquifer, genetic diversity, microbial biomass. S = A·B − P approaches 0. | I: distinguish domain-specific strains. Do not collapse all into "soil decline." Each domain requires its own evidence. |
| 3. Repair Capacity Decline | Θ (repair traction) falls as K (resistance) rises. Compaction prevents biological tillage. Chemical residues disrupt soil biology. Salinization limits plant and microbial capacity. Economic pressure prevents management change. | Θ = σ(−1 + θ_C·C + θ_E·E − θ_K·K). K rises: compaction, salinity, toxicity, debt lock-in. C falls. Θ approaches 0. | H: Θ decline is often invisible in yield data. Must be measured directly (soil biology, infiltration). Do not assume repair capacity from surface appearance. |
| 4. Degradation Propagation | One domain's failure propagates into another. Soil compaction → infiltration failure → drought stress → crop loss → chemical escalation → biology disruption → more compaction. Eutrophication: excess nutrient runoff → aquatic hypoxia → fishery loss. | Ξ = cascade propagation. Λ (coupling strength) high when multiple domains are degraded simultaneously. D grows at growth_per_cycle. | I: cascades require multi-domain tracking. Cannot diagnose Ξ from a single domain measurement. Cascade attribution requires watershed-level evidence. |
| 5. Production Masking | Yield may remain temporarily stable through external inputs (fertilizer substituting for biology, irrigation compensating for water retention loss, pesticides replacing ecological pest control, debt financing marginal operations). Yield looks fine. C is being consumed. | U (yield) maintained by external input while D grows and C → 0. This is the most dangerous OAM state in agriculture: D accumulating while U appears stable. | H: yield data alone is insufficient. Must measure C directly. I: yield stability ≠ system health. This is AG-R-01 in operation. |
| 6. Threshold Crossing | Hidden losses become visible as production collapse, land abandonment, aquifer failure, fishery loss, or food security crisis. The collapse appears sudden but has been accumulating for years or decades in D while C was being silently consumed. | D has accumulated to the point where U can no longer be maintained even with external inputs. S < 0. Decompensation visible. | H: threshold crossing is not predictable from any single metric (AG-UA-09). Threshold must not be claimed without long-term time series. |
| 7. Regeneration Pathway | Repair requires: substrate return, organic matter accumulation, water retention improvement, microbial/fungal support, biodiversity recovery, reduced extraction, diversified production, restored return cycles, and time. ΔD = −β·U·C·L_life·R_s·E·Θ must become > growth_per_cycle. | ΔD > growth: net D decrease. Requires all six factors in ΔD equation to be non-zero. K must fall (compaction break, salt leach, toxicity reduction). L_life = life-first yield orientation active. | I: Regeneration requires time-series evidence (AG-R-13, AG-UA-07, AG-UA-10). Cannot claim regeneration from short-term observations. No prescription. |
| 8. Resonance Restoration | A system becomes resonant when production, soil repair, water security, biodiversity, human stewardship, and return cycles reinforce each other rather than cannibalize one another. Rn > 0 sustained. This is not maximum extraction. It is life-compatible production over time. | Rn = √(Fidelity × Cohesion). Fidelity: honest signal + structural consistency. Cohesion: mutually supporting soil/water/biology/labor/food system. Resonance > 0 = life-compatible agricultural function under pressure. | R: resonance is not a state that can be declared — it must be demonstrated by time-series evidence across multiple domains. Resonance ≠ maximum productivity. |
Agriculture is the planetary-life continuation of the abiogenesis / Big Bloom cycle. Soil is not where food comes from — soil is where life continues to become possible. The return loop is not a management option; it is the condition that makes future agriculture physically possible.
| OAM degradation agriculture | HIR regeneration agriculture |
|---|---|
| Biomass exported; soil nutrients not replaced | Return is measured; substrate loop documented |
| Soil exposed; erosion and oxidation accelerate D | Soil covered; erosion minimized; C protected |
| Water drawn beyond recharge; C (aquifer) depleted | Water retained; infiltration maintained; recharge monitored |
| Soil biology weakened by chemical pressure or compaction | Biology supported; Θ maintained; K reduced where measurable |
| Inputs substituted for biological services; dependency grows | Biodiversity rebuilt; biological services measured where possible |
| Yield maintained temporarily; D invisible in metrics | Yield redefined as life-compatible production; multi-metric health tracked |
| Ecological debt accumulates; future capacity consumed | Return cycles documented; ecological accounting attempted |
Agriculture is where many prior layers become physical — thermodynamics in soil heat and water cycles, biology in the microbial network, ecology in the biodiversity boundary, and the abiogenesis Big Bloom cycle in the return of organic matter to soil substrate.
| Domain | Agricultural analogue | Signal | Structure | Boundary | Pressure | Degradation risk | Repair path | HIR boundary |
|---|---|---|---|---|---|---|---|---|
| Abiogenesis | Soil as ongoing chemical substrate; decomposition as prebiotic recycling; mycorrhizal networks as proto-metabolic exchange | Soil chemistry signals | Soil aggregate / food web | Soil horizon / root zone / watershed | Extraction, tillage, compaction | Biological complexity loss | Organic matter return, root architecture | Soil ≠ inert chemistry. No life-origin claim. |
| Thermodynamics | Solar energy → photosynthesis → biomass → decomposition → heat and nutrients returned. Open-system energy throughput. | Soil temperature, solar radiation | Energy balance per field | Field boundary, watershed | Albedo change, soil carbon loss → CO₂ | Energy balance disruption, carbon efflux > influx | Ground cover, carbon sequestration management | Second law holds. No thermodynamic proof from yield. |
| Ecology | Agricultural ecosystem is a managed subset of natural ecology — same principles apply: diversity, food webs, cycles, thresholds | Biodiversity counts, trophic indicators | Ecological community structure | Habitat edge, buffer zones | Habitat simplification, monoculture | Trophic cascade failure | Biodiversity restoration, buffer habitat | Ecology ≠ agriculture. Do not claim equivalence without data. |
| DNA / Genomics | Genetic diversity in seed banks and crop varieties is agricultural C (reserve). Monoculture = genomic narrowing. Seed sovereignty = genetic HIR. | Variety list, genetic diversity index | Crop genetic architecture | Seed bank, species boundary | Genetic uniformity, IP restrictions | Vulnerability to pathogen (AC-10) | Landrace preservation, wild relative protection | No genomic diagnostic claim. Diversity ≠ performance. |
| Brain / Body Health | Soil microbiome parallels human microbiome. Food nutrient density → human health. Agricultural degradation → food system health burden. | Nutrient density (AG-M-16), food access data | Food system | Farm boundary, watershed, supply chain | Nutritional density loss, food access inequality | Diet-related health burden | Dietary diversity, local food access | No dietary or clinical claim. Food access ≠ dietary outcome. |
| Biofeedback | Precision agriculture sensors are a biofeedback loop for the farm: soil moisture, NDVI, weather data inform management. Same signal/proxy/interpretation caution applies. | Sensor data (AG-M-06, AG-M-20) | Farm monitoring system | Field boundary, sensor coverage | Sensor artifact, calibration failure | Management decision based on bad data | Calibration, ground truth, time series | Sensor = proxy. AG-UB-04, AG-R-05 apply. |
| AI Inference | AI applied to agricultural data faces same overclaim risks: yield optimization ≠ system health; prediction from incomplete data; site-specific advice without local data | Remote sensing, sensor arrays | AI model architecture | Inference boundary, model scope | Hallucination, insufficient local data | Wrong prescription, confidence overclaim | HIR governance, uncertainty taxonomy | AI output ≠ agronomic advice. AG-R-16 applies. |
| Supply Chain | Food supply chain has same fragility vs. resilience tension as AI supply chains: concentration, efficiency vs. redundancy, brittleness | Supply-chain data (AG-M-23) | Distribution network | Supply-chain boundary, geographic | Commodity concentration, just-in-time brittleness | Food access collapse on shock | Redundancy, regional capacity, diversification | Efficiency ≠ resilience (AG-R-15). AG-UB-09. |
| Society | Agricultural labor, food sovereignty, equity of access, and farmer knowledge are social systems embedded in food systems — cannot be separated from ecological analysis | Labor data (AG-M-24), food access (AG-M-26) | Social and economic structure | Community, market, policy | Economic extraction, labor burden, inequality | Knowledge loss, stewardship capacity loss | Labor rights, food sovereignty, equity | No political prescription. AG-R-17 applies. Systems pressure ≠ individual fault. |
// Architecture sample — not site-specific, not prescriptive { "record_id": "AG-SAMPLE-001", "pathway_id": "AG-P-01", "domain_id": ["AG-D-01", "AG-D-03"], "farm_or_region_scope": "field", "location_precision": "regional", "pressure_sources": ["W: high-intensity rainfall events on bare soil; F: continuous row-crop without ground cover over multiple seasons"], "oam_P_estimate": "high", "oam_D_estimate": "moderate-to-severe", "oam_K_estimate": "moderate — established hardpan if repeated compaction co-occurring", "oam_Theta_estimate": "low — bare soil limits biological repair traction", "production_masking_suspected": true, "measurable_signs": [ { "measurement_id": "AG-M-09", "value_or_description": "USLE-estimated erosion rate", "single_point_only": false, "uncertainty_classes": ["AG-UB-03", "AG-UB-05"] }, { "measurement_id": "AG-M-01", "value_or_description": "Declining SOM trend", "single_point_only": false, "uncertainty_classes": ["AG-UB-02", "AG-UB-03"] }, { "measurement_id": "AG-M-20", "value_or_description": "NDVI decline in erosion-prone areas", "uncertainty_classes": ["AG-UB-07"] } ], "biological_unknown_class": ["AG-UA-07", "AG-UA-08"], "measurement_unknown_class": ["AG-UB-03", "AG-UB-05", "AG-UB-12"], "evidence_level": "multi_metric", "confidence_level": "moderate", "HIR_boundary_note": "Erosion rate estimate ≠ actual soil lost. Topsoil depth change requires direct measurement. Recovery timeline is site-specific. No ground-cover or tillage prescription.", "OAM_pathway_note": "D accumulates as topsoil depth declines. C (SOM, biological reserve) falls with topsoil. K rises if hardpan forms. Ξ: erosion → runoff → water quality → downstream eutrophication potential.", "intervention_claim_allowed": false, "site_specific_claim_allowed": false }
{
"record_id": "AG-SAMPLE-002", "pathway_id": "AG-P-03", "domain_id": ["AG-D-01", "AG-D-02", "AG-D-04"],
"pressure_sources": ["F: continuous tillage oxidizing organic matter; residue removal; no organic return; long history of conventional annual cropping"],
"oam_P_estimate": "moderate-high (F dominant)", "oam_D_estimate": "moderate",
"oam_C_estimate": "significantly_reduced — SOM historically high in pre-agricultural profile but consistently declining",
"production_masking_suspected": true,
"measurable_signs": [
{ "measurement_id": "AG-M-02", "value_or_description": "SOM time series showing decline", "single_point_only": false, "uncertainty_classes": ["AG-UB-02", "AG-UB-03"] },
{ "measurement_id": "AG-M-10", "value_or_description": "Declining microbial respiration", "uncertainty_classes": ["AG-UA-01", "AG-UA-02"] }
],
"biological_unknown_class": ["AG-UA-01", "AG-UA-02", "AG-UA-07"],
"evidence_level": "time_series", "regeneration_status": "REVIEW_REQUIRED",
"HIR_boundary_note": "SOM decline ≠ complete soil health claim. Carbon proxy does not capture full biology (AG-R-04). Single-season input response ≠ regeneration (AG-R-07). No tillage or organic matter prescription.",
"OAM_pathway_note": "C (SOM) declines as D grows. Without return loop (AG-D-15), ΔD cannot exceed growth. Θ declines as microbial biomass falls. K rises if compaction co-occurs. Production masked by fertilizer substituting for biological nutrient cycling.",
"intervention_claim_allowed": false, "site_specific_claim_allowed": false
}
{
"record_id": "AG-SAMPLE-003", "pathway_id": "AG-P-07", "domain_id": ["AG-D-03"],
"pressure_sources": ["F: annual irrigation withdrawal rate exceeding aquifer recharge rate over decades; arid to semi-arid climate with low natural recharge"],
"oam_P_estimate": "very_high (F dominant)", "oam_D_estimate": "severe — many aquifers at 30–70% historical level SOURCE_NEEDED by region",
"oam_K_estimate": "very high — geological limits on recharge; no biological repair pathway for fossil aquifer depletion",
"oam_Theta_estimate": "absent for fossil aquifer; low for recharge-limited aquifer",
"production_masking_suspected": true,
"measurable_signs": [
{ "measurement_id": "AG-M-12", "value_or_description": "Water table depth declining trend from monitoring wells", "single_point_only": false, "uncertainty_classes": ["AG-UB-11"] },
{ "measurement_id": "AG-M-13", "value_or_description": "Irrigation volume trend vs. estimated recharge", "uncertainty_classes": ["AG-UB-06", "AG-UB-11"] }
],
"biological_unknown_class": ["AG-UA-03"],
"measurement_unknown_class": ["AG-UB-11", "AG-UB-06"],
"evidence_level": "time_series", "confidence_level": "high (where monitoring data exists)",
"HIR_boundary_note": "Water table decline ≠ collapse timeline. Recharge rate requires separate hydrological study LOCAL_DATA_REQUIRED. No irrigation reduction prescription. Economic and food-access context required before any policy interpretation.",
"intervention_claim_allowed": false, "site_specific_claim_allowed": false
}
{
"record_id": "AG-SAMPLE-004", "pathway_id": "AG-P-10", "domain_id": ["AG-D-05", "AG-D-06", "AG-D-09", "AG-D-13"],
"pressure_sources": ["F: decades of genetic uniformity in commercial varieties; market incentives for single high-yielding variety across large geographic areas"],
"oam_P_estimate": "moderate (F dominant — slow-building)",
"oam_C_estimate": "reduced — genetic diversity reserve depleted by commercial selection; landraces and wild relatives not maintained in production system",
"oam_K_estimate": "high — IP restrictions and supply-chain lock-in make variety diversification economically difficult even when ecologically needed",
"measurable_signs": [
{ "measurement_id": "AG-M-18", "value_or_description": "Variety diversity index — commercial production area", "uncertainty_classes": ["AG-UB-03"] }
],
"biological_unknown_class": ["AG-UA-06", "AG-UA-09"],
"HIR_boundary_note": "Monoculture efficiency ≠ resilience (AG-R-09). Genetic uniformity risk is known but threshold for pathogen spread depends on specific pathogen, climate, and geography. No variety recommendation. No seed prescription.",
"OAM_pathway_note": "C (genetic diversity) low. K (IP + supply-chain lock-in) high. Ξ risk: single pathogen capable of propagating across entire production geography. Classic high-D / high-K / low-Θ state. Production masking: yield appears stable until outbreak occurs.",
"intervention_claim_allowed": false, "site_specific_claim_allowed": false
}
{
"record_id": "AG-SAMPLE-005", "pathway_id": "AG-P-13", "domain_id": ["AG-D-07", "AG-D-05", "AG-D-16"],
"pressure_sources": ["F: habitat loss, pesticide exposure, pathogen load (Varroa, nosema), monofloral landscape (low floral diversity), climate shift in phenology"],
"oam_P_estimate": "high (F dominant — cumulative and multifactorial)",
"oam_C_estimate": "reduced — wild pollinator diversity and habitat connectivity declining",
"oam_K_estimate": "moderate — habitat fragmentation limits corridor function; chemical residues persist in environment",
"measurable_signs": [
{ "measurement_id": "AG-M-19", "value_or_description": "Wild bee abundance and diversity survey", "uncertainty_classes": ["AG-UA-04", "AG-UB-03"] },
{ "measurement_id": "AG-M-17", "value_or_description": "Pesticide application records near pollinator habitat", "uncertainty_classes": ["AG-UB-06"] }
],
"biological_unknown_class": ["AG-UA-04", "AG-UA-06"],
"HIR_boundary_note": "Pollinator decline is multifactorial — no single-cause attribution. Pesticide records are one input, not sole cause. No pesticide recommendation. Pollinator count ≠ crop pollination adequacy without crop-specific assessment.",
"OAM_pathway_note": "Cascade risk (Ξ): pollinator loss → crop pollination failure → yield decline → food access impact (AG-D-14). C (wild pollinator reserve) depleting across landscape. Threshold effect risk (AG-UA-09) — decline may accelerate below critical habitat connectivity.",
"intervention_claim_allowed": false, "site_specific_claim_allowed": false
}
{
"record_id": "AG-SAMPLE-006", "pathway_id": "AG-P-05", "domain_id": ["AG-D-04", "AG-D-03", "AG-D-16"],
"pressure_sources": ["W: high-intensity rainfall events mobilizing surface nutrients; F: chronic excess fertilizer application rate, bare soil windows, impervious drainage, tile drainage concentration"],
"oam_P_estimate": "high (F + W coupling)",
"oam_D_estimate": "severe in receiving water body (aquatic hypoxia, species loss)",
"oam_K_estimate": "high in established hypoxic zones — biological recovery requires sustained nutrient reduction and years to decades",
"measurable_signs": [
{ "measurement_id": "AG-M-04", "value_or_description": "Nitrate in tile drainage or surface runoff", "uncertainty_classes": ["AG-UB-05"] },
{ "measurement_id": "AG-M-18", "value_or_description": "Aquatic biodiversity in receiving water", "uncertainty_classes": ["AG-UB-03"] }
],
"biological_unknown_class": ["AG-UA-03", "AG-UA-09"],
"measurement_unknown_class": ["AG-UB-05", "AG-UB-06"],
"HIR_boundary_note": "Eutrophication attribution requires watershed-level data — cannot infer from single field. Single nitrate measurement ≠ loading estimate. Recovery timeline is long and uncertain. No fertilizer reduction prescription.",
"OAM_pathway_note": "Ξ = nutrient → algal bloom → hypoxia → aquatic food web collapse → fishery loss → food access impact (AG-D-14). K very high in established dead zones (Gulf of Mexico: decades of reduction needed). D in aquatic domain continues while on-farm yield masking continues.",
"intervention_claim_allowed": false, "site_specific_claim_allowed": false
}
Stable domain definitions before any pathway or measurement is added. OAM C/K/Θ assignments per domain. Must define what constitutes reserve, what constitutes resistance to repair, and what the repair pathway looks like in each domain — in bounded, non-prescriptive language.
Each pathway defined with its OAM structure before any measurement or site data is attached. Mechanism status and evidence level set per pathway. Production masking flag defined per pathway before any data is attached.
Each measurement type assigned default uncertainty classes before any measurement values are processed. Remote sensing (AG-M-20) flagged AG-UB-07 by default. Farmer observation (AG-M-22) flagged AG-UB-08 by default. Single soil test flagged AG-UB-02 by default.
Each uncertainty class defined with its confidence impact, suspension rule, and hard-override trigger. Integration with prior layer uncertainty taxonomies where relevant (biofeedback BB classes apply to agricultural sensor data).
Schema formalized with required fields, enum values, and hard defaults. intervention_claim_allowed = false enforced at schema level. site_specific_claim_allowed = false as default. regeneration_confirmed = false until time-series evidence documented.
Rules formalized as typed objects with trigger conditions, governance effects, and override logic. Hard override rules machine-enforced (prescription → override, overclaim → blocked, missing data → unresolved).
Sample records must include: a production-masking case (yield stable, C declining), a time-series supported case (multi-year soil data), a cascade case (multiple Ξ domains), and a hard-override case (prescription attempted and blocked).
Every record validated against: HIR_boundary_note populated, intervention_claim_allowed = false, all uncertainty classes appropriate, confidence level consistent with evidence level, no prohibited inferences present.
Only activates when actual measured data from specific location, time, and management history is documented. Requires soil test records, water monitoring data, management records, and local climate data. Does not activate prescription capability — still requires expert review (Layer 10) for any management inference.
The only layer from which bounded site-specific interpretation (not prescription) can be produced. Requires: documented local data (Layer 9), qualified domain expert review, explicitly bounded output (not a general recommendation), and clear documentation of what was and was not assessed. Still cannot produce pesticide, fertilizer, or irrigation prescriptions through this system.
This layer should not dominate the Rice AI in Health submission. Its appropriate placement is in the broader life-first systems appendix or supplementary materials — demonstrating that HIR × OAM generalizes into structured uncertainty governance and degradation mapping across life-supporting systems beyond the clinical domain.
| Submission component | Agriculture layer positioning |
|---|---|
| Primary Rice technical argument | Absent or one-sentence reference. Primary argument: HIR-governed bounded AI inference, OAM degradation mapping for health systems, genomic uncertainty, biofeedback, pathophysiology, and neuropathology. |
| Health-relevant connection | Food systems → environmental health → public health → dietary diversity → food access as a social determinant of health. These connections can be briefly noted without agricultural prescription claims. |
| Framework generalizability demonstration | Agriculture layer shows that HIR × OAM uncertainty governance and degradation mapping applies to soil, water, and food systems — the same grammar that governs clinical AI applies to environmental AI. One paragraph in appendix context. |
| What to NOT present in Rice materials | Agricultural management advice, specific farming practices, claims about regenerative agriculture as a proven outcome, any prescriptive content. Do not present as medical or agricultural advice of any kind. |