Abstract: A thunderstorm is an important weather factor to operation safety of unmanned aerial vehicles (UAVs).To obtain real-time precipitation data and give the alert information is of great significance for the safe operation of UAVs.Based on registration and projection transformation, a method to calibrate weather radar mosaic is proposed.Furthermore, a dynamic Z-R fitting scheme for radar quantitative precipitation estimation (QPE) is developed.In the end, a method for quantifying threat degrees of precipitation to UAVs is presented to give weather alert accordingly.A case study shows that the proposed methods can be implemented to improve the precision of radar mosaic and the accuracy of radar QPE.Moreover, the dynamic fitting of Z-R is proved to be more accurate than the fixed Z-R for precipitation estimation.The cumulative damage can be used as a reference for risk warning of UAVs.