Temporal and design approaches and yield-weather relationships

The climate changes and the weather events affect agricultural production and farmers’ income. Several strategies may help improving the resilience of farms to climate change, and particular mention should be done to the weather index-based crop insurance schemes, as they rely on the yield-weather relationship. A vast majority of studies investigate the limitation of the weather index insurance, due to the complex relationships linking weather events and yields and the difficulty to capture them with an index (i.e., the basis risk). The literature has not devoted sufficient attention to compare different specifications within the same statistical model in yield weather estimation. Our study, conducted on durum wheat in Italy, shows how the identification (and design) of the phenological stages (i.e., temporal specifications) may help capturing or depicting the yield-weather relationships. The negative effects of the low temperatures, especially during the early stages of durum wheat, is remarkable. Our findings contribute to the debate on the design of triggers in weather indexes (e.g., for minimum temperatures), stimulating new research directions to assist stakeholders interested in planning agricultural risk management interventions.