Abstract: In order to prevent the occurrence of youth drowning, we studied a meteorological warning method for youth drowning risk. By analyzing 135 cases of adolescent drowning in Chongqing, and considering the meteorological factors affecting human comfort and subjective judgment, temperature, relative humidity, hours of sunshine and precipitation are identified as the meteorological indicators of drowning warning. Compared the correlations between the meteorological indicators and the average daily temperature on 5, 10, 15 and 20 days before the drowning incident by linear regression, it is determined that the meteorological factors of the 10 days prior to the drowning event had the largest correlation coefficient with the average daily temperature on the day of the drowning event. By principal component analysis, it was determined that the average daily temperature of the 10 days prior to the drowning event was an important factor in the occurrence of the drowning event, and that the average maximum temperature, average minimum temperature and average relative humidity are the main meteorological factors that influence the occurrence of drowning. Finally, according to the results of principal component analysis, an early warning model of drowning events is established and the risk of drowning is classified. The warning method is validated with examples and the results show that it is feasible.