FAO Uses Airbus Imagery to Estimate Rice
In a desert country such as Afghanistan, food production is highly sensitive to climatic conditions.
For organisations like FAO, assessing the extent of irrigated crops such as rice is critical to calculate food security and anticipate any shortages.
However, determining rice surfaces cannot rely on traditional ground surveys alone, as fields are numerous, often small and scattered across the landscape.
FAO developed a crop calendar, outlining the unique pattern of the rice growth cycle. Based on this calendar, they then gathered satellite imagery over seven provinces, at key dates throughout the rice cycle.
Pléiades and SPOT images were collected before and after the harvest, in order to delineate parcels and establish homogeneous areas:
- Optimise sampling, minimise ground surveys and maximise efficiency
- Perform rice classification, with additional time-series based on Sentinel 1, Sentinel 2 and Landsat 8 earlier in the season.
As a result, the rice crop area was successfully estimated with less than a 10% error rate. They even achieved an accuracy of 96% in one particular province.
The use of our satellite imagery, together with complementary free data sources allowed FAO to substantially reduce production costs of crops statistics, while maintaining the highest quality standards:
- Increased reliability of the produced statistics
- Reduced risk for ground surveying
- Improved cost-efficiency.
See our products in action
Related pages
- Pléiades Neo: The best of very-high resolution optical imagery
- Pléiades: Spot the detail thanks to our very-high resolution satellite imagery
- SPOT: The ideal solution for country-wide, demanding applications
- Agriculture market
- Institutional Agriculture: Supporting the modernisation of agricultural policies with timely and accurate geospatial insights
- Crop Analytics for Institutions: Our Crop Analytics for institutions delivers time series of satellite vegetation maps
- Basemap: When the stakes are high, start with the right foundation