Abstract: The pomelo industry is a pillar industry in Pinghe County, Zhangzhou City, Fujian Province. In order to disperse the operational risks of local pomelo production, based on the pomelo production in Pinghe County of Zhangzhou City from 1985 to 2022 and the weather data from the middle ten days March to the first ten days of August of the corresponding year, the original data of pomelo yield per unit area was decomposed into trend yield per unit area and meteorological yield per unit area. The original data of yield per unit area was adjusted to convert it to the yield per unit area under the productivity level of 2024. The kernel density estimation method was used to fit the probability density function of the sequence and used for the determination of pure rate. Then, the correlation analysis was carried out by using the adjusted yield per unit area and weather data to screen the significant weather index. The results showed that the linear relationship between the rainfall index and the adjusted yield per unit area was the most significant in the middle and last ten days of April. For the agreed insurance output of 51000 kg·hm⁻² and the insurance price of 2.8 yuan·kg⁻¹ in 2024, that was, the insurance amount per 666.7 m² was 9520 yuan, the determined pure premium rate was 4.0191%, and the insurance rate was 4.6220%. For the insurance compensation trigger value of the rainfall index RI of 160mm, and the eight graded compensation intervals of RI, the corresponding probability of occurrence for the compensation events was calculated respectively, and the corresponding calculation formula of compensation amount was given.