Abstract: Based on daily and hourly meteorological data, layered soil moisture data and disaster data, the dry-hot wind disasters of winter wheat accompanying high temperature and low humidity are studied in North China and Huanghuai Region through historical disaster inversion, normal distribution test, soil moisture treatment and independent t-test. These disaster samples are classified:Class A samples are not affected by the relative humidity of soil, while Class B samples are affected by the relative humidity of soil. Thresholds of soil relative humidity on dry-hot wind disaster of wheat are determined according to independence of samples in two groups, and verified by random samples. Relative humidity values of the whole and layered soil layers subject to normal distribution. The relative humidity of Sample A and Sample B in each soil layer is independent. Thresholds of relative humidity affecting dry-hot wind disaster are 58%-65% for the whole soil layer and 56%-75% for layered soil. The mean value of the whole layer is approximately 60% and increases with the depth of the soil layer. The coincidence rate of relative humidity threshold of each soil layer is between 72.5% and 85%with average value about 80%, which could reasonably reflect the influence of soil relative humidity on dry-hot wind disaster of wheat. For convenience in application, 60% of the soil relative humidity in depth of 10-20 cm layer is selected as the critical threshold to determine the influence of soil relative humidity on the dry-hot wind disaster of northern winter wheat, being significant when it is greater than or equal to 60% and ignoral when less than 60%. The conformance rate of independent samples is 82.5%. Results provide a scientific basis for quantifying effects of soil relative humidity on the dry-hot wind disaster of winter wheat.