National statistical offices (NSOs) are responsible for producing official agricultural statistics, yet most still rely on ground-based surveys that are slow, expensive, and difficult to scale. Increasingly, NSOs are also expected to contribute to sustainability monitoring frameworks, including SDG reporting, climate reporting and land-use assessments. This narrative shows how satellite Earth observation is being integrated into NSO workflows to support crop area mapping, land use reporting, agricultural statistics and sustainability indicators such as SDG 2.4.1 (Proportion of agricultural area under productive and sustainable agriculture) EO is becoming part of a broader multi-source statistical production system combining surveys, administrative records, geospatial information and satellite observations. NSOs must produce official agricultural statistics under growing pressure to deliver more data, more frequently, with flat or declining budgets.
Key challenges include:
- Survey costs: traditional crop area surveys rely on field enumerators visiting farms across large territories. In many countries, a full agricultural census is conducted once every ten years, and annual surveys cover only a fraction of the total agricultural area
- Temporal gaps: annual or biannual surveys cannot capture in-season crop developments. Governments, food agencies, and commodity markets increasingly need timely data that field surveys cannot provide
- Spatial coverage: sampling frameworks designed for large commercial farms systematically undercount smallholder plots, which dominate agricultural systems in much of the world
- International reporting obligations: NSOs must contribute data to FAO, Eurostat, and the UN SDG monitoring framework. Indicators including SDG 2.4.1 (proportion of agricultural area under productive and sustainable agriculture) and SDG 15.3.1 (land degradation neutrality) require information omn agricultural extend, land cover, vegetation condition and productivity that many NSOs do not currently have the capacity to produce
- Methodology inconsistency: national survey definitions of crop types, agricultural area, and land use often differ from international standards, making cross-country comparisons and global aggregations unreliable
How EO can help
Satellites observe agricultural land across an entire country at consistent intervals, independent of survey budget or field capacity. EO does not replace the ground truth that field surveys provide, but it extends coverage, fills temporal gaps, and reduces the cost of meeting international reporting requirements.
- Crop type mapping: Sentinel-2 time-series at 10 m resolution can distinguish crop types across a full agricultural territory, including smallholder plots that are systematically absent from sample surveys. This directly supports area estimates and SDG indicator production
- Agricultural area estimation: land cover classifications from Copernicus data provide a consistent annual baseline for tracking changes in cropland extent, fallow land, and permanent crops without a full field survey cycle
- Seasonal monitoring: vegetation indices (NDVI, EVI) and Sentinel-1 radar backscatter track crop development throughout the growing season, supporting in-year yield forecasts and early warning of production shortfalls
- Monitoring agricultural sustainability: O can support several dimensions of agricultural sustainability. Time-series observations allow monitoring of vegetation productivity, drought exposure, irrigation practices, soil cover, crop rotation patterns, land degradation processes and ecosystem pressures. These datasets can complement survey-based information and provide consistent national baselines for assessing sustainable agricultural systems, particularly under SDG 2.4.1.
- Standardisation: global EO datasets apply consistent classification methods across national borders, making it possible for NSOs to produce statistics that align with FAO and Eurostat definitions
Key examples
EO as part of the national agricultural statistical infrastructure is increasingly aligned with international efforts to modernise agricultural statistics through integrated multi-source production systems.
1- Sen4Stat
Sen4Stat is an ESA project that provides NSOs with an open-source processing system for producing agricultural statistics from Sentinel satellite data. The system uses Sentinel-1, Sentinel-2, and Landsat time-series to generate crop area and crop type products that can be run locally or on the cloud. The project was developed through direct collaboration with statistical offices in Spain, Ecuador, Senegal, and Tanzania, covering both data-rich environments and countries where ground survey capacity is limited. The system is designed to fit into existing NSO data workflows rather than require entirely new infrastructure.
2- WorldCereal
WorldCereal is an ESA-funded initiative that produces global, seasonally updated maps of cropland extent, crop type, and irrigation at 10 m resolution. The initial product set covers maize, winter cereals, spring cereals, and temporary crops worldwide for 2021 and is freely available. FAO uses WorldCereal data to fill gaps in its global agricultural statistics where national survey data is missing or unreliable. Since December 2023, the project has entered a second phase focused on operational uptake by national agencies and international organisations.
Figure: Winter Cereals 2021 data from World Cereal, integrated in APEX Geospatial Explorer for Statistics
3- FAO-EOSTAT
FAO-EOSTAT is a FAO capacity-building programme that uses satellite imagery and AI to help NSOs in lower-income countries produce standardised land cover maps and agricultural statistics. Launched in 2019 with pilots in Senegal and Uganda, the programme has expanded to 21 countries. It uses freely available Copernicus data to generate annual land cover maps that NSOs can use for SDG reporting. In Lesotho, FAO-EOSTAT has enabled the Bureau of Statistics to report annually on SDG 15.4.2 (Mountain Green Cover Index), a target previously not measurable from national ground data.
4- SEF Statistics APEx Explorer
The SEF Statistics APEx Explorer (explorer.sef-statistics.apex.esa.int) is a live geospatial platform that brings together EO-derived agricultural datasets in a single interface. The Agricultural Production tab includes global cereal maps from WorldCereal (spring and winter cereals, maize, and temporary crops at 10 m resolution for 2021), annual European crop type maps from the Copernicus Land Monitoring Service HRL Cropland product (2017-2023, 10 m resolution, minimum mapping unit 0.25 ha), and on-demand national crop maps produced through the GTIF Green Meridian system, illustrated with a UK crop map for 2022.
Figure: SEF Statistics APEx Explorer
Further resources
CLMS HRL Cropland: Copernicus annual European crop type maps (2017-2023)
