Appendix B — Data Sources, Update Schedules, and Key Data Providers
Geospatial intelligence relies on consistent, reliable, and frequently updated data sources.
Oracle Agriculture Intelligence integrates satellite observations, weather datasets, soil information, terrain models, and geographic reference data to build a comprehensive view of agricultural conditions.
This appendix summarizes the primary data providers used in the platform, their update frequencies, and what each dataset contributes to agricultural monitoring.
Core Observation Data
Sentinel-2 Satellite Imagery
Provider: European Space Agency (ESA) / Copernicus Programme
Mission Purpose: Land monitoring, vegetation analysis, and environmental assessment
Spatial Resolution: 10 meters (visible and near-infrared bands)
Revisit Frequency: Approximately every 5 days at the equator (more frequent at mid-latitudes)
Spectral Bands: 13 bands across visible, near-infrared, and shortwave-infrared wavelengths
What Sentinel-2 provides for agriculture
- High-resolution snapshots of crop conditions
- Vegetation indices (NDVI, EVI, NDWI)
- Crop development and phenology signals
- Detection of vegetation stress
- Land-use classification
- Detection of flooding, drought stress, and bare soil
Why Sentinel-2 is ideal for agronomic intelligence
- Free and openly accessible
- Frequent global coverage
- Multi-spectral observations optimized for vegetation monitoring
- Historical archives enabling multi-year seasonal comparisons
Open-Meteo Weather Data
Provider: Open-Meteo (https://open-meteo.com)
Data Type: Global weather forecasts and historical reanalysis
Coverage: Worldwide, uniform spatial grid
Temporal Resolution: Hourly to daily
Update Frequency
- Forecasts: Updated several times per day
- Historical reanalysis: Periodically updated as models improve
Weather variables used in agricultural intelligence
- Temperature
- Precipitation
- Wind speed and gusts
- Solar radiation
- Humidity
- Soil moisture (model-derived)
- Evapotranspiration estimates
Why Open-Meteo is useful for food security
- Free and openly accessible
- Consistent global coverage for national-scale monitoring
- Enables early detection of weather-related agricultural risks
- Supports both early warning and retrospective analysis
Supporting Environmental Datasets
Soil Composition Data (ISDA / ISRIC)
Providers:
- ISDA (Innovations for Soil and Agricultural Data)
- ISRIC – World Soil Information
Data Provided
Global soil datasets describing physical and chemical soil properties, including:
- Soil texture and structure
- Soil organic carbon
- Soil pH
- Nutrient composition
- Soil depth and layering
Role in agricultural intelligence
These datasets support:
- Crop suitability analysis
- Soil health assessment
- Yield modeling
- Agricultural planning and land management
HydroSHEDS Hydrological Data
Provider: World Wildlife Fund (WWF) and partners
Dataset: HydroSHEDS (Hydrological data and maps based on SHuttle Elevation Derivatives at multiple Scales)
Data Provided
- River networks
- Watershed boundaries
- Drainage basins
- Flow accumulation
- Flow direction
Role in agricultural intelligence
HydroSHEDS supports:
- Flood risk detection
- Watershed analysis
- Runoff modeling
- Water-flow and drainage analysis
These datasets help identify areas where rainfall may accumulate and threaten agricultural production.
Copernicus Digital Elevation Model (DEM)
Provider: Copernicus Programme / European Space Agency
Data Provided
Global terrain elevation data describing the Earth’s surface.
Key characteristics
- High-resolution elevation grids
- Global coverage
- Derived terrain attributes such as:
- slope
- aspect
- terrain curvature
Role in agricultural intelligence
Elevation models support:
- Flood risk analysis
- Hydrological flow modeling
- Watershed delineation
- Identification of erosion-prone landscapes
Geographic Reference Data
Oracle Spatial Reference Layers
Provider: Oracle Spatial and external geographic datasets
Data Provided
Geographic reference layers used to organize and interpret agricultural data.
These include:
- Administrative boundaries (national, provincial, district)
- Infrastructure networks (roads, transport corridors)
- Population centers and settlements
- Geographic units used for reporting and aggregation
Role in agricultural intelligence
These layers support:
- Aggregation of crop and weather insights into administrative units
- Identification of population exposure to agricultural risks
- Analysis of infrastructure proximity and accessibility
- Spatial organization of agricultural intelligence dashboards
Data Update Schedules
The platform ingests satellite and weather data on regular schedules.
Satellite Imagery Updates (Sentinel-2)
| Component | Frequency | Notes |
|---|---|---|
| Raw image availability | ~5 days | More frequent at higher latitudes |
| Cloud-free composites | Weekly or biweekly | Depends on cloud patterns |
| Vegetation-condition time series | Every new image | Supports near-real-time monitoring |
Weather Data Updates (Open-Meteo)
| Component | Frequency | Notes |
|---|---|---|
| Forecast weather | Multiple times per day | Supports emerging risk detection |
| Historical reanalysis | Periodically updated | Supports climate and trend studies |
| Daily indicators | Daily | Used for crop-growth models |
| Extreme event indicators | Daily or sub-daily | Detects anomalies quickly |
Additional Supporting Data
Crop Calendars and Agronomic Reference Data
Provides crop-specific information including:
- Planting windows
- Harvest periods
- Typical crop development stages
This information helps interpret satellite observations within the correct seasonal context.
Historical Production Data
Historical agricultural production statistics provide a baseline for:
- Yield forecasting
- Seasonal comparisons
- Detection of abnormal production years
- Agricultural planning and policy analysis
Field Boundaries or Parcel Maps (Future)
Where available, field boundaries allow more precise agricultural analysis.
These datasets enable:
- Field-level monitoring
- Precision agriculture advisory services
- Farm-level yield estimation
However, parcel data is not required for national-scale food-security monitoring.
How These Data Sources Work Together
Oracle Agriculture Intelligence integrates multiple geospatial datasets into a unified analytical system.
Each data source contributes a different type of information:
- Satellite imagery reveals what is happening on the ground: crop health, vegetation vigor, and land use.
- Weather data explains environmental drivers such as rainfall deficits, heat stress, or storm events.
- Soil and terrain datasets describe underlying environmental conditions that influence crop productivity and water movement.
- Hydrology datasets help detect flood risk and watershed dynamics.
- Geographic reference layers organize insights into meaningful reporting units for governments.
Together, these datasets enable continuous monitoring of agricultural landscapes and provide early insight into potential food-security risks.