Contaminant Build-Up in Urban Snow Cover

Aleksandar Đukić1, Branislava Lekić1, Vladana Rajaković Ognjanović1


1 University of Belgrade – Faculty of Civil Engineering, Kralja Aleksandra 73, 11000 Belgrade, Serbia; E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it



The focus of this paper was the study of short-term accumulation of contaminants in surface layers of snow cover in urban areas. Within the paper the results of analysis of pH value, conductivity, solid matter, organic matter, nutrients, sodium, potrassium and anions concentration in snow cover samples are presented. Samples were taken from a parking lot in downtown Belgrade, Serbia, over a 7-day antecedent dry period during winter. Results exhibited significant variability. Build-up of most of the measured parameters were best described by power functions. The highest surface loads were observed for solids and chemical oxygen demand. All samples were mildly acidic and conductivity was low. After the beginning of snow melting, a significant drop of anion concentation and an increase in solids content were observed, while pH value increased to neutral.

Keywords: contaminant, solids, anions, build up, snow, urban areas.


Pollution build up in urban areas is a complex process consisting of deposition of substances from the atmosphere, deposition of wind-blown particles and accumulation of materials on the surface due to human activities (transport, industry, etc.). Thus, amounts of dust, pollution and other substances in urban areas increase over time (Hvitved-Jacobsen et al. 2010).

Pollution in the atmosphere usually exists in the form of small particles or attached to particle surfaces. The largest deposition of atmospheric particles takes place in one of the following ways (Hvitved-Jacobsen et al. 2010):

  • Dry deposition that occurs during dry weather conditions, where sedimentation and adsorption processes are responsible for the accumulation of substances in urban areas,
  • Wet deposition which takes place by absorption of particles in raindrops during their transport through the atmosphere,
  • Occult deposition which takes place by absorption of pollutants in the small droplets in clouds and fog.

In addition, water droplets in clouds and precipitation can absorb gases present in the atmosphere, such as carbon dioxide, ammonia, sulphur dioxide and others (Hvitved-Jacobsen et al. 2010). The relative distribution between the solid, liquid, and gaseous phases greatly affects the basic characteristics of pollution in terms of their transport, transformation and effects. Terms of balance and dynamic characteristics in the transport of pollutants between the gaseous and particulate phases, are essential factors in understanding the aspects of the quality of surface runoff from urban areas (Hvitved-Jacobsen et al. 2010).

Snow in urban areas has a high potential to accumulate pollution: Glenn D.W.III, Sansalone J.J. (2002) reported that urban roadway snow exposed to traffic and winter maintenance practices has a much greater capacity to accumulate and retain pollutants than storm water runoff, while event mean concentrations of all particle sizes are reported to be higher during the melt period compared to the rain period (Westerlund and Wiklander, 2006). Also, Deletic and Orr (2005) reported that sediment loading and anions concentration on road surfaces were highest in the winter months, especially when snow was present on road surfaces. For pollutant removal from snowmelt and runoff, sorptive filtration systems are being developed (Liu at al. 2005).

The focus of this paper is the study of short-term accumulation of contaminants in surface layers of snow cover in urban areas. This may be of particular importance for assessing pollution accumulation in snow and pollution release during snow melt in urban areas with a moderate or continental climate during the winter months, where winter weather is characterised by alternating cold and warm periods where snow forms and melts several times during the winter.


Materials and Methods

The study site is located in an urban environment in downtown Belgrade, Serbia, within the parking lot of the University of Belgrade - Faculty of Civil Engineering. The study site catchment area covers 921 m2 and is comprised of 74% impervious surfaces (asphalt, concrete) and 26% green areas. The experimental site is located 50 m from the surrounding streets and is separated from heavy traffic by a 0.7 m high concrete wall and a line of planted trees (Djukic at al. 2016). Snow sampling was carried out on December 10, 12, 13, 14, 15 and 17 of 2013. Snow cover formed during December 8 and during the period between December 8-15, air temperature was below 0°C. On December 16 and 17, air temperatures rose and snowmelt began, and sampling of snow was no longer possible after December 17.

Six samples of top snow layers, weighing approximately 1 kg each, were obtained from a 0.25 m2 area of undisturbed snow cover. Snow samples were only collected from undisturbed snow cover, each was collected from a different area with the site. Collected samples were placed in plastic bags and sent to the laboratory where they were melted and transferred to plastic bottles and stored under refrigeration at 4°C until the laboratory analyses were performed. A list of the examined parameters and applied analytical methods are given in Table 1.




The most frequently used functions, which describe the process of pollution accumulation on surfaces over dry periods, are described by equations (1-4) (Hvitved-Jacobsen et al., 2010; Rossman 2015).

Linear (LIN) function means that surface contamination increases over time as described by equation (1):

for01             (1)


B – accumulated pollutant mass at the surface at time t (mass / area)

Bo – initial mass of pollutant accumulated (mass/area)

C2b – pollutant build up rate (mass/area/time)

t – antecedent dry weather period.

It is assumed that atmospheric pollution, from traffic and other sources, is continuously being deposited on surfaces. However, studies have shown that although daily accumulation of pollution always exists, the pollution accumulation rate usually decreases over time, leading to a slowdown in surface pollutant accumulation and saturation, corresponding to surface type, surface use and other factors (Hvitved-Jacobsen et al., 2010, Rossman 2015, Wicke et al. 2012). The most commonly used functions to describe pollution accumulation slow down over dry weather periods are given by equations (2-4) (Hvitved-Jacobsen et al., 2010, Rossman 2015).

Power (POW) function:

for02                     (2)


C1b - the maximum pollutant build up on the surface (mass/area),

C2b – pollutant build up rate.

Exponential (EXP) function:

for03                      (3)

where B and C1b have the same meaning as in the previous equation, and C2b – coefficient of removing the pollutants due to various factors (time-1).

Saturation (SAT) function (Michaelis-Menten):

for04                           (4)

where C2b is half-saturation constant for the given pollutant (days).

When determining the appropriate functions that describe the accumulation of certain pollutants in each examined case, thebest determination is obtained by both, field and laboratory research. It should be taken into account that the coefficient values in the accumulation equations may vary widely and may differ by an order of magnitude (Wicke et al. 2012; Rossman 2015).


Results and Discussion

After measurements, samples were neutral to mildly acidic, with a pH value ranging between 4.8 and 6.9 (Figure 1), which is consistent with previous studies (Glenn and Sansalone, 2002). The highest pH value was measured in the last sample. Electrical conductivity values were very low (Figure 1). The hardness of measured samples was also very low, ranging from 4 to 11 mgCaCO3/L.


Figure 1: Electrical conductivity and pH in analysed snow samples.


Sodium and potassium-ions were not detected in any of the samples, while phosphates ions were detected only in two samples with values of 0.172 and 0.192 mg/L. Obtained results for other parameters in samples taken during the December between 10th and 15th, are presented in Figure 2. Results are given in units of mass of pollutant per unit area (kg/ha). It can be observed that results show high variability. Samples taken on December 13 and 14 have lower levels of some of the analzyedparameters, which may indicate the dependence of the measurement results on the sampling location. With the exception of the results for these two samples, a trend of increasing values for all sample parameters was observed.

The mass of suspended solids and chemical oxygen demand in snow are almost two orders of magnitude lower than the mass of pollutants accumulated on asphalt surfaces from the same area during summer and autumn (Djukic et al. 2016). This can be explained by the short sampling period and the fact that the snow cover was fresh (on the first day of sampling the snow cover was two days old). Also, air quality may be another contributing factor, but available data on air quality is not sufficient to make conclusions in this regard.

Results of snow sampling analyses were used to determine which of the build up functions (POW, EXP, SAT and LIN) best describes the results. For the LIN function it was assumed that the initial amounts of pollutants were zero, and in POW function C1b was assumed to be zero (the maximum pollutant build up on the surface). The coefficients C1b, C2b and C3b of build-up functions were determined by minimizing the sum of the squared differences between the modelled and experimental results (R2) using Excel Solver (Microsoft Excel, 2007), and are presented in Table 2.




Figure 2: Measured values in snow samples and build up functions POW, EXP, SAT and LIN


The power function best describes build up of almost all analysed parameters (Table 1). The power function for TSS and COD even has a concave increasing shape (Figure 2), which can be explained by the very high ability of snow to bind pollutants (particles). An increasing trend for TS, TSS and COD is more noticeable than for other parameters, but a decrease in build-up rate was not clearly indicated for any of the investigated parameters, except for fluorides (Figure 2).

Earlier studies have shown that snow is very effective in binding the particles which are later released into the environment only after thawing (Glenn and Sansalone 2002, Westerlund and Wiklander 2006). On December 16th and 17th snow was melting and analyses of snow samples taken on December 17th show a significant change in pollutant accumulated: TS increased significantly from 2.52 to 4.20 and TSS increased from 1.37 to 3.80 kg/ha, while COD slightly decreased from 1.96 to 1.59 kg/ha. This indicates that, even while melting, snow continues to accumulate particles, and at the same time compacting of the surface snow layers may further increase concentration of pollutants in the top layers. A slight decrease of COD may be explained by migration of soluble COD fraction from snow into the snowmelt runoff.

All anions show a significant decrease in snow sample taken on December 17: fluorides 0.00076 kg/s, Chlorides 0.042 kg/a, nitrates 0.0128 kg/ha, nitrites 0.0014, sulphates 0.0323 kg/ha. This can also be explained by migration of these soluble substances from snow into the water (snow melt runoff). A significant drop of electrical conductivity in this sample (Figure 1) can also be attributed to the reduction of the ion concentration of snow.



The results of a study of different physico-chemical parameters in snow cover samples taken from a parking lot in downtown Belgrade, Serbia. Samples were taken over a 7-day antecedent dry period during winter. The results show that accumulation of pollutants in snow is much lower that on asphalt surface on the same location during dry weather periods in summer and autumn. Results also exhibited significant variability and build-up of most of the measured parameters was best described by power functions. The most prominent increasing trend shows TS and TSS, the values which continue to increase even during snowmelt. From the results it can be concluded that saturation of surface layers of snow with examined pollutants was not observed during observed 7 days antecedent dry weather period, except for fluorides. It can be also concluded that, even while melting, snow continues to accumulate particles, and at the same time the surface snow layers may further increase concentration of pollutants in the top layers.



The authors are grateful to the Serbian Ministry of Education, Science and Technological Development for its financial support (Projects No. TR37009, TR37010).



DeletIc A., Orr D. (2005) Pollution buildup on road surfaces. Journal of Environmental Engineering 131 (1), 49-59

Đukić A., B. Lekić, Rajaković-Ognjanović V., F. Veljovic, Vulic, T., M. Đolić, Naunović Z. Despotović J., Prodanovic D., (2016), Further Insight into the Mechanism of Heavy Metals Partitioning and Stormwater Runoff, Journal of Environmental Management, 168, 104-110

Glenn D.W.III, Sansalone J.J. (2002) Accretion and Partitioning of Heavy Metals Associated with Snow Exposed to Urban Traffic Storm and Winter Maintenance Activities. II. Journal of Environmental Engineering 128, 167-185

Hvitved-Jacobsen T., Vollertsen J., A. Nielsen (2010) Urban and Highway Stormwater Pollution-concepts and Engineering. CRC Press. Taylor & Francis Group, Boca Raton, FL, USA.

Liu D., Sansalone J.J., Cartledge F.K. (2005) Adsorption Kinetics for Urban Rainfall-Runoff Metals by Composite Oxide-Coated Polymeric Media. Journal of Environmental Engineering, 131(8), 1168–1177

Rossman L.A. (2015) Storm Water Management Model User's Manual Version 5.1, EPA 600 / R-14 / 413b, U.S. Environmental Protection AGNECY

Westerlund C., Wiklander M. (2006) Particles and associated metals in road runoff during snowmelt and rainfall. Science of the Total Environment 362 (2006) 143- 156

Wicke D., T. A. Cochrane, A. O'Sullivan (2012) Build-up dynamics of heavy metals deposited on impermeable urban surfaces. Journal of Environmental Management 113(2012), 347-354