An early warning system for wheat production in low rainfall areas of Jordan

The objective of this study was to construct forecasting models using climatic parameters such as accumulated rainfall, average monthly temperatures and cultivated area, which could predict, at an early stage of the growing season, the expected level of domestic wheat production in low rainfall areas. The translog function showed the best prediction capability among different models. The model was constructed based on past records of the accumulated rainfall during the growing season up to December and up to January. The results indicated that it is possible to predict wheat production in low rainfall area using only the rainfall, temperature and area data available in the earlier months of the growing season. Analysis showed that rainfall was the major factor in increasing wheat production. Interestingly, yield predictions made at the end of January, i.e. including the rainfall in January, are no better than those made at the end of December. An additional 10-mm precipitated in December or January will increase wheat production by 6•2% and 13•1%, respectively. As these predictions can be made at the end of December, the decision-makers should have sufficient time to draw-up the necessary programme of imports.