Agriculture + Technology + Finance
Agro Capital Standard brings land, climate, satellite and market data into the analysis behind agricultural finance, investment and project development.
Understand the land → Examine the evidence → Inform the decision
Agro Capital Standards is the knowledge and methodology hub of the ACS ecosystem. Explore the data, products and analytical principles behind land due diligence, agricultural financing, carbon projects and ongoing monitoring.
Connect land and crop context with due diligence, risk analysis and financing design. Explore the roles of TerraScore, ACS Bank OS and the Corn / Oilseeds Capital initiatives.
Explore financial applicationsUnderstand how field information, climate and soil evidence support farm planning and carbon-project preparation through ACS Farmer, TerraScore and CARBON.
Explore CARBON projectsExplore data provenance, analytical methods and human oversight, including the roles of ProvableCORE, SentientROUTER and SoulGCI in governed technology.
Explore the technologyAgricultural TechFin for banks, insurers and investment funds across the United States, Argentina and Brazil — combining ERA5 climate, Sentinel-2 NDVI, USDA, IBGE and CONAB data, provenance, risk analytics and financial decision support.
Evidence is data with a stated source, period, quality and scope.
NO DATA → NO SCORE
The dashboard exposes a coverage status per region, per crop and per risk surface for six analytical surfaces: risk_score, CVI, capital, regime, satellite_summary and NDVI. These states are component-specific; consult the returned output:
This turns data gaps into visible governance signals. Bank credit committees can see which surfaces support an assessment. Data integrity is a first-class output.
Provenance records origin, transformations, versions and traceability.
Risk analysis interprets evidence under a defined methodology and its limitations. Financial decision support provides material for an authorized human or institutional decision.
What happens if drought hits during flowering? If corn prices drop 30% at harvest? If the local currency devalues overnight? If all three happen at once?
The Monte Carlo methodology combines yield shocks, price movements, climate extremes and currency changes. Inputs can include Copernicus ERA5, USDA and World Bank data, with periods and weights defined per calculation.
The result is a full probability distribution — P10 (worst realistic), P50 (expected), P90 (best realistic) — for IRR, DSCR, and loss probability.
Computation time depends on workload and configuration.
"Other platforms give you a credit score. We describe the scenarios behind it."
Built on 4 Rust high-performance engines: Monte Carlo · Risk Score · NDVI Processor · ERA5 Parser. Zero Python in the hot path.
Field assessment, climate and satellite data, soil evidence and project preparation for independent verification. CARBON connects farmers, landowners and project partners, with an initial focus on Europe. Start with your location, crops and farm area.
Framework roles connect agricultural evidence, institutional review, financial decision support, capital workflows and monitoring. Capital and financing proceed through authorized contractual and institutional processes.
Framework role: evidence → institutional review → financial decision support.
Framework role: evidence → institutional review → financial decision support.
Framework role: farm/operator context, identity and primary evidence.
Framework role: commodity evidence → buyer/value-chain context → financing design → monitoring.
Framework role: oilseed evidence → buyer/value-chain context → financing design → monitoring.
Framework role: land evidence → provenance → due diligence.
Framework role: provenance, evidence receipts and verification.
Framework role: policy-based routing and approval gates.
Framework role: field, soil and climate evidence → project preparation → independent verification.
Framework role: governance and human authority for AI workflows.
ProvableCORE is the AI governance layer intended for supported ACS integrations. It is designed to generate cryptographic evidence for supported decision paths — verifiable by banks independently, without access to ACS infrastructure.
EU AI Act Article 12 addresses record-keeping for in-scope high-risk AI systems; applicability requires legal review. ProvableCORE delivers tamper-evident audit receipts designed to support these compliance reviews. Bank credit committee remains the final decision-maker.
Data foundations include agricultural observations and market and environmental inputs. Monitoring feeds new observations and events back into analysis.
Risk Methodologies and Financial Decision Support
These six named surfaces form the 227,902,135-row core baseline. The public estate metric covers 248,982,733 rows across 103 current nonempty analytical tables in 16 analytical datasets.
Counts reflect current BigQuery table metadata. The analytical estate excludes rollback and snapshot copies, backups, staging, duplicates, test and CI fixtures, billing, and superseded versions; provenance and observation periods remain source-specific.
Copernicus CDS · Google Earth Engine · USDA NASS · IBGE SIDRA · FAO STAT · NASA POWER · World Bank · ECB · Eurostat · ISRIC SoilGrids
AI-assisted workflows connect agricultural context, portfolio analysis, insurance signals and supply-chain evidence with human and institutional review.
Subscription, model usage and API charges are defined for each engagement.
The geographic registry organizes country, crop and evidence context across European, Black Sea and Americas evaluation tracks.
Farmer identity, consent, agronomic history and operator evidence are organized for institution-led assessment.
Farmer ID and HCI provide a framework for farm and operator context, identity and primary evidence.
"Your ACS ID is what stays with you when everything else changes."
Collateral-led review with limited operator history and agronomic context.
Documented operator profile combining identity, agronomic history, yield evidence and portability context.
Security, governance and regulatory reference controls connect technical evidence with institutional review.
Governance defines authority, review, exceptions and accountability throughout the chain.
| Feature | Traditional Credit Scoring | Agri-Weather Platforms | ACS |
|---|---|---|---|
| Basel II PD/LGD/EAD concepts International banking standard for measuring credit risk | Established practice | Adjacent | Risk framework |
| Satellite vegetation evidence Sentinel-2 NDVI provides vegetation-condition evidence | External input | Core capability | Evidence input |
| Monte Carlo portfolio simulation Configurable "what-if" scenarios stress-testing drought, price, and FX shocks | Portfolio method | Scenario input | Risk methodology |
| Human Capital Index (farmer rating) Operator context: discipline, efficiency and adaptivity | Institution-specific | Adjacent | Operator framework |
| Portable Farmer ID across banks UUID identity designed around the farmer | Institution-specific | Adjacent | Identity framework |
| SHAP explainability per decision Transparent breakdown of each risk factor's contribution | Model-specific | Model-specific | Explainability framework |
| Covenant monitoring workflows Alert design for loan-condition events | Bank workflow | Signal input | Monitoring framework |
| Regional benchmarks with coverage status Peer comparison across the Americas evaluation track, with AVAILABLE / PARTIAL / INSUFFICIENT / BLOCKED state shown per region and crop | Portfolio-specific | Core capability | Coverage framework |
| Credit committee evidence packs Structured evidence for authorized institutional review | Bank workflow | Data input | Decision-support framework |
| Output-specific provenance Source, period, geography and lineage remain attached to analytical outputs | Institution-specific | Source-specific | Evidence framework |
ACS Bank OS uses controlled individual access keys for qualifying banks, insurers and investment funds. Each engagement begins with a scoped data-coverage review for the institution's geography and crop mix, using AVAILABLE / PARTIAL / INSUFFICIENT / BLOCKED states. Americas regions follow the data and evaluation track, connected to the European Bank OS framework.
To request access, contact contact@agrocapitalstandard.eu.
The primary presentation metrics cover 44 countries, 248M+ analytical observations (248,982,733 rows across 103 current nonempty analytical tables in 16 analytical datasets), and 26 complete climate years from 2000–2025. The analytical total excludes rollback and snapshot copies, backups, staging, duplicates, test and CI fixtures, billing, and superseded versions. The six named tables in Agricultural Data Foundations remain a 227,902,135-row core baseline.
Technology metrics use the configured 10M Monte Carlo path capacity, 467 mounted API routes in the current authorization inventory, 5,080 passing automated tests in the current acs-api-python main CI suite, and 41 farmer tools in the current product catalog.
Data provenance, geography, model validation and product scope are output-specific. Simulations are model outputs. ACS provides analytical decision support, while binding credit and investment decisions remain with authorized people and institutions. Regulatory applicability, certification, contractual service levels and product-specific terms are established through the relevant formal process.
Current repository, CI and data-plane evidence for the public presentation
