Skip to main content

Data Model & Ontology

AgriFoodData serialises the digital farm twin as JSON-LD, with semantic anchoring in AgroVoc (FAO) and the Web of Things vocabularies.

Why JSON-LD​

  • It is plain JSON — every existing client library reads it.
  • It carries an explicit @context, so any field name maps unambiguously to a URI in an open vocabulary. This makes data interpretable across systems and languages.

Example payload — Field with AgroVoc tags​

{
  "@context": "https://agrifooddata.org/contexts/farm-v1.jsonld",
  "@type": "Field",
  "id": "fld-0042",
  "name": "North field",
  "farm": "farm-0001",
  "geometry": {
    "type": "Polygon",
    "coordinates": [[[78.49, 17.39], [78.50, 17.39],
                     [78.50, 17.40], [78.49, 17.40],
                     [78.49, 17.39]]]
  },
  "agrovoc": [
    "http://aims.fao.org/aos/agrovoc/c_7156",   // maize
    "http://aims.fao.org/aos/agrovoc/c_7156-soil-clay"
  ]
}

AgroVoc in practice​

  • Use AgroVoc URIs for crops, soils, management practices, inputs and outputs wherever a free-text field would otherwise drift.
  • The Farm API exposes /agrovoc/lookup so you can resolve labels in any AgroVoc-supported language without bundling the full vocabulary.

See it in action​