The improvement of methodologies for crop yield forecasting has been identified as one of the central objective of the AMIS project “Strengthening Agricultural Market Information Systems globally and in selected countries using innovative methods and digital technology” funded by the Bill and Melinda Gates Foundation. Crop forecasting at sub-national, national and international level is fundamental to provide decision makers and private actors with timely information for rapid and sound decisions during the growing season. Crop production and yield forecasts are also key data for making food policy decisions. Almost all major food security programmes rely on crop estimations and forecasts for strategic planning. For these and other reasons, strengthening crop production forecasting constitutes a core research priority of AMIS.
In practice, crop production forecasts are obtained as the product of two components (1) the estimation of area devoted to a given crop, and 2) the estimation of expected yield per unit of area. While there are no major constraints with the available methodologies for area estimation, forecasting yield is still a major challenge in many developed and developing countries. In order to fill existing methodological gaps with yield forecasting, following activities are implemented:
- Review and document existing practices for crop yield forecasting, selecting countries with recognized expertise in this area and at different stages of statistical capacity development. Besides the methodological issues, this review focuses on human and physical resources needed to set up, run and update the system. Particular attention is devoted to yield forecasting methodologies suitable for the AMIS core crops (wheat, maize, rice and soybeans).
- Identify yield forecasting methods using a combination of ground and remotely sensed data, looking at the required data infrastructure and institutional set-up behind each method.
- Provide guidance on the identified trade-offs between cost/capacity/quality related to the implementation of each method. This research is conducted in close cooperation with the Global Strategy to Improve Agricultural and Rural Statistics (GSARS).