Abstract: Weather index insurance with the advantages of transparent information, convenient claim and settlement strong secondary market liquidity is an effective method to diversify agricultural risks, compared with traditional agricultural insurance products. However, the disaster mechanism of crops that are often affected by multiple hazards together during their growth process is very complex. Accurately constructing the correlation relationship model among multiple weather indices and between weather index and yield per unit area of agricultural products, expanding the underwriting liability of weather index insurance are important to reduce the basis risk, reasonably design weather index insurance and transfer the risk of agricultural weather disasters. In this study, based on the precipitation and temperature data of 24 years of day-by-day from 1995 to 2018 from the national meteorological network, a three-dimensional nested Copula model and a conditional mixed three-dimensional Copula model among precipitation, temperature and yield per unit area had been constructed for kiwifruit in Meixian county, Shaanxi province. The relationship between the three variables was simulated using the conditional mixed three-dimensional Copula model with higher simulation accuracy by comparing the error and the pure rate of composite weather index insurance was determined. The results showed that the pure insurance rate was 10.07% at 70% coverage level for a payout under the conditions of cumulative precipitation below 423.2mm and average daily maximum temperature above 26.40℃ from May to September. This study explored the cross-influence of high temperature and drought weather on the unit yield loss of kiwifruit and better clarified the correlation among high temperature, drought and kiwifruit yields, which could reduce the basis risk to a certain extent, improve the agricultural weather index insurance system, provide a new idea and method for the design of composite weather index insurance products, and contribute to the promotion and application of weather index insurance.

Key words: Cross-influence, Copula function, Monte Carlo simulation, Rate determination