The invention discloses a thunder and lightning prediction method. The method comprises the following steps: obtaining basic meteorological parameters of a to-be-predicted area; calculating high-ordermeteorological parameters related to thunder and lightning based on the high-order meteorological parameters of the to-be-predicted area; acquiring lightning positioning observation data of the to-be-predicted area, and performing gridding processing on the lightning positioning observation data; based on a random forest algorithm, calculating the degree of correlation between each high-order meteorological parameter and thunder and lightning, and selecting the high-order meteorological parameter with the highest degree of correlation with thunder and lightning; establishing a forecasting model by utilizing an XGBoost algorithm based on the forecasting timeliness and the forecasting time; and based on the high-order meteorological parameters of the to-be-predicted area, predicting the spatial distribution and occurrence probability of thunder and lightning by utilizing the prediction model.