Information on prospective crop yields over large regions is essential for government services dealing with the import and export of food crops, agencies with a role in food relief, international organisations with a mandate for monitoring world food production and trade, and commodity traders.
For example, organising food aid missions requires information on where and when shortages and surpluses are expected to occur.
The Food and Agriculture Organisation’s ARTEMIS system for real-time crop and vegetation monitoring in Africa is one of several existing agricultural information systems. Another system is the crop explorer developed by the United States Department of Agriculture, whose crop monitoring service is tailored more towards commodity trading. A similar but independent system is available in Europe. Although this is designed primarily for European conditions, it is currently being extended to obtain global coverage.
European MARS crop yield forecasting system
In Europe early information on crop yield and production is vital for policy making within the framework of the Common Agricultural Policy. The Centre for Geo-information is the lead partner in a consortium providing agricultural crop monitoring services to the Joint Research Centre of the European Commission. This monitoring system, currently known as the MARS Crop Yield Forecasting System, comprises three main tasks: providing information on weather, which is carried out by MeteoConsult; processing and visualising results from satellite imagery, which is carried out by the Flemish Institute for Technology (VITO); and monitoring conditions for crop growth and yield forecasting, which is carried out by Alterra.
Crop models are important aids in monitoring crop growth conditions across Europe. The WOFOST crop model (WOrld FOod STudies) was developed in the 1980s to simulate crop growth at point locations using daily weather variables and to predict crop yield and other crop related variables. Within the MARS system WOFOST is applied throughout Europe on a 50x50 km grid to simulate the growth and yield of 12 important crops on a daily basis. Time series of crop simulation results are then aggregated to a national level and used as an input to a statistical analysis which forecasts the current year’s crop yield. The spatial application of WOFOST is better known as the Crop Growth Monitoring System (CGMS).
Global Crop Water Stress Index
In 2005 efforts were made to develop a global system for monitoring crop conditions based on the Crop Water Satisfaction Index (CWSI). The concept of CWSI was developed by FAO and is much simpler than the WOFOST model, which makes it attractive to implement on a global scale. The weather data required to calculate the CWSI are derived from global weather forecasting models and are available with a resolution of 1 degree.
New programming techniques have made it possible to create an interactive interface for querying the CWSI database. This interface allows the user to navigate quickly through the calculated CWSI values for different crops, make comparisons with other years or long-term average values and analyse the CWSI for specific locations using detailed calculation reports.
The Geoland project (EU 6th Framework Programme) examines the accuracy of the yield forecasts. Five different systems for yield forecasting developed by different European partners are being assessed at three test sites (Poland, Belgium and Spain). Results for Spain obtained with different variants of CGMS demonstrate that the accuracy of the forecasts increases during the crop growth season. The red line is the accuracy level that is acceptable by the end user and the chart demonstrates that some of the variants can meet this criteria two months before the end of the season. The information gathered within the Geoland project can be used to improve the yield forecasting system performance by selecting the most accurate variants.