So, what are the factors that can modify the temperature structure in the lowest few hundred metres in marginal snow situations?
1. SYNOPTIC-SCALE MODIFICATION of the temperature structure of the lower troposphere. If the air passes over the sea (or similarly warm surface), then the sensible flux of heat to the air above will raise the ZDL, perhaps tipping the balance towards rain or sleet, rather than snow - windward coastal plains may miss out on the worst of the snow. ( However, these same areas may be the only places to experience moist convection in winter and provided the air is cold enough, and the sea is close and upwind, then snow showers can be frequent. ) Heat from major urban areas (provided areally extensive) can also tip the balance in highly marginal situations. If the air passes over an ice or snow-covered surface, then a flux of heat from the air to the surface occurs, modifying the ZDL structure, usually resulting in a sharp, shallow inversion. The air-mass (highest altitude) ZDL is unlikely to be affected but a secondary pair of ZDLs may form as the thermal structure of the lowest 300m is distorted and either freezing rain or ice pellets, rather than 'proper' snow is the result. This is often a difficult situation to get right after a long cold period is trying to break down.
2. EVAPORATIVE COOLING of the air through which the snowflakes are falling. Even with the most intense precipitation, there is always lots of air around the falling raindrops or snowflakes and evaporation of the precipitation elements will occur. This will lead to a microscale cooling (due to latent heat exchanges as the liquid/ice evaporates), which multiplied by the huge number of precipitation elements leads to a net cooling of the environment through which the droplets/crystals are falling. This in turn leads to a lowering of the ZDL. The effect is proportional to the precipitation intensity ( and inversely proportional to the mean wind speed through the melting layer ) and is greater when the ambient relative humidity is well under 95%.
3. PHASE-CHANGE COOLING of the air through which the snow is falling. Rain/snow situations are often marginal at low altitudes. This means that more often than not, snow is melting in the lowest 200m or so and thus the environment is cooled due to heat exchanges consequent upon the melting of the ice crystals into liquid water. Again, intensity of precipitation is a major factor - greater intensity means that there are more precipitation elements involved which means greater overall cooling. The effect compounds that at (2) above, the net effect of evaporative and phase-change cooling is quite significant - lowering the ZDL by some hundreds of metres in prolonged precipitation. This is especially pronounced in stable air and catastrophic in near-isothermal conditions in a frontal zone. Some of the worst low-level icing conditions for aircraft occur in these situations, and of course, the ground isn't very far away!
4. BULK (DOWNWARDS) ADVECTION of cold air due to drag by precipitation elements and by downdraughts in a markedly convective environment. Another effect that is related to precipitation intensity is the cold air that is dragged down by the falling elements and the associated downdraughts. Descending air warms adiabatically so this introduction of colder air from upper levels is offset somewhat and is the least effective modulator of those considered above. [ However, in such situations, the relative humidity will fall (greater separation between air and dew point temperature) so evaporative cooling will become more effective - see (2) above. ]
5. OROGRAPHIC UPLIFT COOLING. As air in a thermally stable environment is forced to rise over a range of hills or mountains then the adiabatic cooling will cause the temperature to fall with height more rapidly than in the undisturbed environment. This will lower the ZDL allowing a greater downward penetration of the snow that might otherwise be expected. (This is important in a few of our major towns and cities that rise into the 'foothills' of major hill ranges, e.g. Manchester, Sheffield & Bradford.)
6. FALLING OR SETTLING PROBLEM. Apart from the factors mentioned at the end of (1) above, snow is pretty well guaranteed to fall and settle if the surface temperature is at or below 0 degC. Rain is almost guaranteed if the surface temperature is above 4 degC. In between there is a degree of uncertainty and quite small changes in intensity can switch between sleet and snow, and between snow thawing faster than it falls or vice-versa.