The Winter Precipitation Type Research Multi-scale Experiment (WINTRE-MIX) was designed to study the multi-scale processes influencing the variability and predictability of p-type (rain, drizzle, freezing rain, freezing drizzle, wet snow, ice pellets, and snow) and amount under near-freezing surface conditions. The experiment was conducted from February-March 2022 near the U.S. / Canadian border, in the vicinity of Montreal. The field campaign utilized operational networks (New York State Mesonet, Canada Foundation of Innovation Climate Sentinels) and research instruments like the NRC Convair-580 research aircraft with a suite of in-situ and remote sensors, three ground based-radars: one C-band on Wheels (COW) and two X-band Doppler on Wheels (DOWs) radars, five mobile sounding systems, and four manual p-type observation stations. This presentation will analyze the variability of mesoscale cloud and precipitation structures and mesoscale flow during near-freezing conditions using ground-based radars. Small-scale vertical motions within clouds (convective generating cells, coherent wave motions, and shear-driven turbulence) are shown to enhance the formation of ice in supercooled clouds, leading to enhancement of surface snow or rain. Mesoscale precipitation bands, produced by either convergence of mesoscale terrain-channeled flows or by embedded disturbances within synoptic storms, locally enhance vertical motion, increase cloud depth, and intensify precipitation rates. Consequently, these bands favor transitions towards ice-phase precipitation at the surface via enhancements to low-level diabatic cooling, to initiation of ice aloft, and to riming and aggregation that lead to deeper melting layers.