Abstract: In this study, for the five cases of snowfall in the Nanjing area from December 2023 to February 2024, we analyzed the evolution characteristics and correlations of the microphysical quantities of snowfall particles. Additionally, we established the relationship between snow density (ρ_s) and snowfall rate (SR) estimations using minute-resolution observations made by a PARSIVEL² disdrometer and a weighing precipitation gauge. The results showed that the differences in the microphysical characteristics of snowflakes were closely associated with temperature–humidity transition layers, and that the formation of large snowflakes was mainly attributable to melting and aggregation processes induced by inversion and warm layers. The peak diameters, peak number concentrations, and mean spectral width of the snowfall particle size distribution increased with increasing SR; these trends were consistent with the gradual decrease in the mean values of the shape parameter µ and slope parameter Λ obtained from Gamma function fitting. The distribution of the µ–Λ fitting relationship for different types of snowfall was relatively concentrated, especially at Λ<20 mm⁻¹, which is similar to the results of a study in the Colorado region. Moreover, the mean values of µ and Λ for dry snow were lower than those for wet snow; however, the two µ–Λ fits were consistent with each other. Additionally, the ρ_s–D₀ relationship (ρ_s=0.36D₀^−1.34) and ice water content (IWC)–SR relationship (SR=4.23·IWC) were fitted to analyze the contribution of microphysical quantities (normalized intercept parameter N_w, median volume diameter D₀, and IWC) to different SRs. The SR contours show significant differences in the microphysical properties of snow under the same SR conditions, with SR mainly associated with low concentrations of large-particle snowfall or high concentrations of small-particle snowfall. Finally, the relationship between the equivalent radar reflectivity factor (Z_e) and SR was established as Z_e=173·SR^1.54. The variation in the prefactor a of this relationship was analyzed, and a theoretical expression was proposed. Comparison results showed that lower snowflake terminal velocities V_t and smaller values of the intercept parameter N₀ caused an increase in a.