Snow deposited in urban areas is exposed to pollutants originating from traffic, wet and dry atmospheric deposition, de-icing chemicals and traction materials. The conventional pollutants found in urban snow include solids (TSS), metals (Zn, Cu, Cd, Pb, etc.), Polycyclic Aromatic Hydrocarbons (PAHs), and chlorides. Microplastics (MPs) are emerging pollutants of interest and their presence in urban snow was reported only recently. The pollutants accumulate in snow deposits over time and may be released when snow melts, and the resulting snowmelt may carry pollutants to the receiving waters. Understanding the concentrations and mass loads of pollutants in snow and pollutant behaviour during snowmelt is helpful for planning and developing site-specific snow management practices. The overall aim of this thesis was to evaluate the quality of snow in urban areas, with respect to: (i) quantity of metals, PAHs and MPs in urban snow storage piles and roadside snowbanks, (ii) compare the quality of snow collected in the same catchment during repeated sampling campaigns and evaluate the effect of the sampling design on estimation of pollutant loads in snow piles, and (iii) investigate the pollutant release patterns and temporal variations in their concentrations in water leaving the melting snow piles (in laboratory). For such studies, snow samples were collected from snow storage piles in Frihamnen (a port facility in Stockholm, Sweden) and roadside snowbanks in Luleå and Umeå cities in Northern Sweden.

The quality of snow collected in the three study areas varied considerably, because of differences in such area characteristics as the annual precipitation and snowfall, the population, average daily traffic, land use activities, and snow management activities. The average values of major parameters in analysed snow samples were as follows: TSS - 1500 mg/L, conductivity- 2.1 mS/cm, Zn – 870 µg/L, Cu – 240 µg/L, Cd – 0.48 µg/L, Cr – 120 µg/L, Pb – 50 µg/L and the sum of 16 PAHs – 3.5 µg/L. Microplastics were abundant in urban snow samples, with the following descending order of concentrations: black road wear particles, consisting of bitumen and tire wear particles,  mean = 19300 ± 47400 particles/L; road marking paints with the mean of 430 ± 998 particles /L; and, plastics particles, mean 33 ± 34 particles /L. No correlations were found between the numbers of MP particles and the site-specific parameters.

Comparison of snow pile sampling designs revealed that systematic 1-m square grid sampling yielded the best estimates of mass loads (BEML) of pollutants, compared to single snow cores, or horizontally composed core samples. The mass loads estimated from composite or single snow column sampling deviated up to 50 and 400%, respectively, from BEML.

Results of the laboratory snow melting indicated that PAHs in the snow samples were mostly attached to the particles; only 10% of the total PAHs burden was contributed by the meltwater and the rest stayed on the ground with the sediment residue. The dissolved concentrations of PAHs were below the detection limit (0.010 µg/L) in all the analysed samples except for Fluoranthene and Pyrene with concentrations ranging between 0.01 and 0.02 µg/L. PAHs displayed a delayed release from snow piles, which was similar to that of TSS. Truly dissolved fractions (<3000 MWCO, Molecular Weight Cutoff) of Zn, Cu and Cd represented 71-90% of dissolved fractions in the snow samples collected in Luleå (snow without road salt) and 74-98% in those from Umeå (snow with added road salt). Both dissolved and truly dissolved metals showed advanced releases from all the snow piles. The influence of road salt on releases of metals and PAHs from laboratory snow piles was hard to discern, because of great differences in snow quality characteristics at both locations.