Contaminant Buildup and Distribution on Urban Impervious Surfaces at Parking Lots - page 2

Results and Discussion

The local climate conditions data were obtained from the nearest weather monitoring station. The observed time span was 30 days. Parameters that were recorded were: daily temperatures (max, min, average), wind speed, humidity and precipitation and are presented in Fig. 3.

Four rain events, of which only two with significant precipitation (13 and 11 mm/day), occurred during the last week of July 2012. The sampling campaign began two days after the last rain event in July, when the impervious surfaces being examined appeared to be completely dry. The weather between the first and the second sampling campaigns was extremely hot with average daily temperatures ranging between 25-30oC, while the maximum daily values reached 39oC. Hot weather conditions, combined with an increased wind speed and an absence of additional precipitation caused a sudden drop of air humidity which remained at nearly 40 % for the next 3 weeks until the end of the third sampling campaign.

The temporal accumulation of TSS and heavy HM from the experimental catchment, during the 17 day antecedent period is presented in Fig. 4.


Figure 3: Weather conditions before and during the sampling campaigns.


Figure 4: Temporal accumulation of TSS and HM at different sampling sites.


Both of the sampling sites (A1 and B1, as shown in Fig. 1) that were examined for temporal accumulation of contaminants show a significant decrease in contaminant loads between the first two sampling campaigns. The significant differences in TSS and HM surface loads between the first and second sampling campaigns may be attributed to traffic intensity and weather conditions. A significant decrease in the number of cars using the parking lot during August and a general decrease in traffic intensity in Belgrade during the summer (due to summer holidays) caused a significant decrease in the rate of contaminant buildup. Additionally, light-to-moderate rain events that preceded the first sampling campaign did not produce significant wash-off of accumulated material from surfaces, which may have accumulated in significant amounts, as the parking lot is used with a much higher frequency during the month of July (especially in the first half of the month). Besides the two previously described rain events in the second half of July, there was an additional rain event recorded on July 7 that amounted to 13 mm/day. As reported by Zhang et al. (2012) a significant portion of pollutants can to some extent be retained on the surface after lighter rain events.

The trend of a slight decrease and stagnation between the second and the third sampling campaign was observed for location A1 while a decrease in the contaminant load continued for location B1. High loads of TSS for both examined locations during the first sampling campaign are attributed to the condition of the 10 year old asphalt surface and the absence of mechanical street-sweeping in the studied parking area. The same trend is noticed for iron and zinc in terms of the load and accumulation rate. Changes of surface loads and accumulation rates are moderate for the other examined heavy metals, with a slight observable declining trend. It is assumed that the declining trend of contaminant accumulation in time is enhanced by local atmospheric conditions, the sporadic usage of the parking lot during the study period, and low traffic intensity in the surrounding streets. Particulate matter is weakly bonded to the asphalt surface under the direct influence of increased daily temperatures, low humidity and increased wind speed. Furthermore, the dried accumulated contaminants tend to erode and dislocate from the experimental surface in hot and dry weather conditions and increased wind speed.

The changes in surface load and the heavy metals accumulation rate follow the same trend as observed for TSS in Fig. 4. This is a rough indicator that heavy metals are potentially bonded to suspended particles in urban runoff. This assumption was further investigated by applying Pearson's correlations coefficients between TSS and heavy metals, as well as for other investigated parameters, the results of which are presented in Table 1.

The linear correlation coefficient (or Pearson's correlation coefficient) measures the strength and the direction of a linear relationship between two variables. A correlation greater than 0.7 is generally described as strong, whereas a correlation less than 0.5 is generally described as weak (Živojinović et al., 2013).

Oils and grease, TN, BOD5 and Cl- are weakly correlated to other urban runoff pollutants indicating high solubility of these chemical species in urban runoff and/or the absence of interaction with analyzed runoff contaminants.

TSS and TS are strongly correlated with TP and HM. This is attributed to physical or chemical bonding of contaminants to the particulate mater, and therefore TSS can be considered as a surrogate parameter for particulate matter in urban runoff. Such small particles are assumed to bond/adsorb higher quantities of contaminants as they have a relatively large surface area per unit mass and therefore a higher adsorption rate, and they have lower densities and contain a greater proportion of organics and certain clay minerals that make them adsorb metals easier. However, the correlation between SS and HM was found to be very weak which, to some degree, supports the conclusion that HM are not predominantly bound to coarse particles. Strong correlations are also observed between TP and HM, particularly iron. Iron is not generally considered a contaminant. However, as a constituent that typically occurs in relatively high concentrations in stormwater runoff, iron may play a role as a bonding agent, in a precipitated and solid state, that adsorbs other pollutants (Eriksson et al., 2005). Removal of heavy metals from stormwater is enhanced by the processes of chemical co-precipitation and aggregation of sorbed material.