VITO Remote Sensing is an experienced international player in the use of satellite imagery for global agricultural monitoring. On behalf of the ESA, it mapped all agricultural plots on Earth using the WorldCereal tool. The tool is a gamechanger in monitoring food security.
WorldCereal is an innovative and dynamic tool developed by an international team under the leadership of VITO. WorldCereal offers the first-ever cereal and maize maps based on satellite imagery, on a global scale and in unprecedented detail. Using freely available satellite data from the EU's Copernicus programme, WorldCereal can update the crop map of any region of the world one month after each agricultural season. This update is rapid: if, for example, the wheat and maize season in western Europe ends in September, the updated map can be ready as early as October. And it also happens automatically, based on global crop growth calendars: when a maize or wheat season ends in a given region, the analysis of that season starts.
The WorldCereal team has also identified where irrigation is essential for successful agriculture. This is crucial information about water use in these times of climate change. The WorldCereal tool is therefore ideal for monitoring food security.
Food speculation and climate change
WorldCereal provides crucial information that allows international organisations such as the Food and Agriculture Organisation of the United Nations (FAO) to enhance and update global agricultural statistics. For example, in cooperation with NASA, the system was already used in 2022 to objectively map agricultural production in Ukraine.
There are a broad range of potential applications for the WorldCereal tool. An organisation like FAO, for example, could use it to counter speculation on food crops, by releasing objective data on their availability (e.g. harvest forecasts) in good time. It could also be useful in the fight against global warming. The WorldCereal maps can help in monitoring variations in greenhouse gas emissions between different types of agriculture and crop types - something that can help countries in drawing up their mandatory 'CO2 accounts'.
One of VITO's strong points that was particularly useful in the development of the WorldCereal tool was the identification and classification of crops based on satellite images. The experts at VITO developed methods to identify winter and spring wheat, for example, and to differentiate it from other cereals. The algorithms for this have been extensively tested and improved in recent years, in the context of various ESA and EU projects.
Of course, the WorldCereal tool was also extensively tested in its entirety before it was released. An important aspect in this regard were tests on a large scale, which included demonstrations of WorldCereal products in five different countries across three continents. These demonstration regions were selected in collaboration with users and based on elements such as agricultural type, plot size and average cloud cover. But the importance of countries in global cereal production was also looked at, including regions with reduced food security. The selected countries were Argentina, France, Ukraine, Spain and Tanzania. In total, the demonstrations covered 1.5 million square kilometres of agricultural land. The results of the monitoring were compared with reference data that had not yet been used during the development of the WorldCereal tool. The comparison showed how accurate the monitoring was.
For the five countries combined, the total annual area of agricultural land was mapped with an accuracy of almost 90%. In France and Ukraine, the monitoring of individual products was the most accurate (more than 95%), while in Spain and Tanzania it was somewhat lower. In these countries, there were overestimates of the extent of agricultural fields, because they were sometimes mistaken for natural vegetation such as grassland.
The next steps
WorldCereal is an open source and scalable tool. As a result, not only national and international institutions, but also the private sector, such as insurance companies, can capitalise on this innovative tool. For example, for monitoring agricultural damage more effectively after extreme weather events.
Although VITO, as the developer, and ESA, as the project financier, have now handed over the tool, there are possibilities for a follow-up phase. ''The user will most likely need support in using the tool,' explains Sven Gilliams of VITO. ''Of course, as developers, we are ideally placed for that.'' In addition, the tool can also be improved and further refined, for example to identify even better and more crop types. This follow-up research and development are of course also on VITO's radar.
