In cold climate regions, seasonal snow accumulation in road ditches plays a critical role in governing the thermal behavior of road embankments. Snow exerts a spatially uneven thermal influence on heat transfer within the embankment due to its strong insulating properties, which can lead to differential frost heave and the formation of cracks on the road surface. However, accurately quantifying this insulation effect requires detailed knowledge of the thermal regime of the snow cover accumulated in the ditches, including its temporal evolution throughout the winter season.

Snow is a porous medium characterized by a high volume fraction of air, which significantly reduces the thermal conductivity of snow and provides a strong insulating capacity. This insulation creates a natural temperature gradient between the snow surface and its base, governing the rate of heat transfer through the snow. Once snow is deposited, a temperature gradient develops, initiating metamorphic processes that alter the morphology and bonding of the snow grains. These microstructural changes lead to temporal variations in the thermal conductivity of snow, making its accurate determination over time particularly challenging. Furthermore, freezing-melting cycles modify the content of unfrozen liquid water within the snow structure, which not only complicates the quantification of latent heat exchanges but also exerts a significant influence on the effective thermal conductivity.

To address these challenges and gain a deeper understanding of the thermal dynamics in snow-covered road embankments, two experimental field sites were established in Luleå, Sweden. The first site, located on the premises of the Luleå University of Technology, was designed to investigate the thermal regime of a natural undisturbed snow cover and its impact on temperatures at the soil–snow interface, as well as within the shallow subsurface soil layer. The second site was located in a road ditch near Luleå Airport, where the snow cover consisted of a mixture of naturally accumulated snow and plowed snow from the adjacent road. The purpose of this site was to investigate the thermal regime of the snow cover in the ditch and its influence on the thermal conditions of the road embankment. The objective was to investigate the thermal regime of both the snow cover in the ditch and the road embankment influenced by snow accumulation during the winter season.

At both sites, the thermal regime of the accumulated snow was continuously monitored by measuring the snow temperature at different heights throughout the winter season. Meteorological conditions were also recorded using a weather station installed at each site. At the site with natural snow accumulation, the thermal influence of snow cover on subsurface soil was measured down to a depth of 40 cm. In contrast, at the experimental site in the road ditch, the thermal regime of the entire road embankment was investigated. To provide a comprehensive overview, the temperature distribution was measured from the road surface to a depth of 2 m, as well as from the surface of the ditch to the same depth.

Analysis of the thermal regimes of snow cover at each experimental site reveals that the natural undisturbed snow cover possesses a greater insulating capacity compared to snow cover accumulated in ditches, due to differences in metamorphic processes. The metamorphism of each type of snow cover is influenced by factors such as history of deposition and environmental conditions.

To assess the insulating behavior of the snow in the ditch, the thermal conductivity of the snow cover at different heights was evaluated over time. Since manual measurements are restricted to discrete points, estimation of the thermal conductivity of snow over a period enables the consideration of continuous changes in the properties of snow driven by metamorphism. To achieve a realistic estimate, a mathematical approach was applied to analyze heat transfer processes within the snow cover based on field measurements. For depths where melting occurred, a novel method was developed to estimate the unfrozen water content of the snow, a parameter that is otherwise very difficult to measure or determine. Based on the analysis, an empirical relation was proposed to estimate the thermal conductivity of snow in the ditch, which is practically useful under similar climatic conditions. However, it was not possible to derive an empirical formula for the unfrozen water content, since the snow undergoes different metamorphic processes at different depths, resulting in highly variable behavior.

Furthermore, the thermal regime of the road embankment during the winter season was assessed using a one-dimensional finite-difference model validated against field measurements. In this model, a new approach was introduced to account for cloud cover in calculating incoming long-wave radiation, which plays a key role in the energy balance of the road surface and strongly affects the surface temperature. The influence of traffic was also incorporated by deriving a representative daily traffic load based on measurements collected over a two-week period. In general, the model improved the estimation of the road surface temperature and provided a more accurate assessment of the thermal regime of the road embankment.

The findings of this study improve the understanding of the thermal regime of road embankments in typical cold-climate regions with significant snow precipitation. Incorporating the thermal behavior of snow in design and maintenance strategies offers critical guidance for implementing measures that mitigate freezing-induced damage and improve the long-term durability of embankments. Furthermore, understanding the thermal influence of snow cover in road ditches can support the development of targeted snow management strategies to optimize its insulating effect. Such strategies may include the complete or partial removal of snow from the ditch or densification of the snowpack to modify its thermal impact on the road embankment. These measures can help improve the long-term performance of the embankment.