The characteristics of a snowstorm event in Changsha of China in 2021 are analyzed based on multi-source observations. Results show that before the snowstorm, the dry layer disappeared, the moist layer thickened, and the wet-bulb temperature was below 0 ℃. The saturated stratification in the cold wedge below the inversion layer was conducive to the aggregation growth of falling snowflakes. Analysis of the wind profile radar data shows a close correlation between the intermittent light rain (snow) in Changsha on December 25 and the passage of several short-wave troughs. The temporary weakening of snowfall on December 26 is attributed to the transition from southwesterly wind to westerly wind. Besides, the correlation coefficient of 0.9–0.97 and uniform differential reflectivity are the characteristics of snowfall echoes. The analysis of the laser raindrop disdrometer data reveals that in the rain stage, particles had high falling velocity and small diameter, and the droplet spectrum was narrow. In the sleet stage, the droplet spectrum broadened rapidly, with an increase in particle number and diameter. In the snow stage, particles had lower falling velocity and larger diameter, and the droplet spectrum was broader. Additionally, the average droplet spectrum distribution (single peak pattern) of all stages is very close to that in the snow stage. The number concentration of raindrops decreases rapidly as the particle diameter increases. Overall, the multi-source new-type observation data with high spatio-temporal resolutions are effective in phased precipitation forecasts.