Bridge Deck Runoff Water Quality Modeling - The "Gazela" Bridge Case Study

Katarina Krstić1, Aleksandar Đukić2, Željko Vasilić2

 

1 Delta Inženjering, Zaplanjska 86, Belgrade; E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

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

 

Abstract

In order to examine pollution emissions washed off by surface runoff from urban roads, a rainfall-runoff and pollutant load model of the bridge deck of the Gazela Bridge in Belgrade,covering an area of 13600 m2 was developed. Runoff and pollutant load modeling was performed using the EPA SVMM software package. Two types of rainfall data were used for rainfall-runoff analyses: uniform rainfall and measured rainfall data for the period from 24 March to 23 August 2014. The model results indicate that the greatest pollutant load from the analyzed impervious surfaces is washed off by runoff resulting from the first 10 mm of rainfall. Various road sweeping strategies were analyzed together with their impacts on the total amounts of washed-off pollutants, and recommendations for sweeping efficiency and frequency are defined.

Keywords: bridge deck, runoff, pollution, street sweeping, dynamic wave.

Introduction

 

Particles carrying various pollutants accumulate on road surfaces during dry weather conditions. These particles are separated and washed off from the surface by runoff during rain events. A schematic representation of changes in the total amount of pollution accumulated on a surface over a time span covering both dry and rainy periods is given in Figure 1 (Hvitved-Jacobsen et al., 2010)

 

fig01
Figure 1: A schematic representation of changes in the total amount of pollution accumulated on impervious surfaces over time (Hvitved-Jacobsen et al., 2010).

 

The accumulation of substances on urban and road surfaces is a complex process consisting of atmospheric deposition, deposition by wind and accumulation of surface pollutants generated by human activity (e.g. vehicular traffic, etc.). Roads are a source of particles larger than 6 μm having high content of heavy metals, with these particles laying on terrain surfaces at distances of up to 150 m from the roads (Sabin et al., 2006). Atmospheric pollution is usually in the form of fine particles or matter adsorbed onto the surface of particles (Hvitved-Jacobsen et al., 2010). In addition, there are also processes that remove pollution from impermeable surfaces: wind, decomposition, street sweeping, etc. (Hvitved-Jacobsen et al. 2010, James et al. 2010).

Described complexity of interactions is the main reason why surface pollutant accumulation is modeled exclusively by empirical relations which most often express the change in pollutant mass per unit area B during time t in one of the following four ways (Bertrand-Krajewski 2007, Hvitved-Jacobsen et al. 2010):

Linear function (LIN):

for01      (1)

Power function (POW):

for02     (2)

Exponential function (EXP):

for03     (3)

Saturation function (SAT):

for04     (4)

Where: B is the pollutant build-up over a time period t (mass/area); t, Δt is the time or time period, C1b is the maximum possible build-up (mass/area), C2b is the build-up rate constant or half-saturation constant and C3b is the time exponent.

Accumulated pollutants are washed off during precipitation when runoff carries accumulated particles and sediments from the surface. Pollutant wash-off is usually modeled as Exponential Wash-off (EXP) or Rating Curve Wash-off (RTC). Exponential wash-off assumes that the wash-off load (W), in units of (mass/time), is proportional to the product of runoff raised to the wash-off exponent C2w and to the remaining amount B (James et al. 2010, Rosmann 2015):

for05     (5)

where C1w – is the wash-off coefficient, C2w – is the wash-off exponent, q – is the runoff rate per unit area and B – is the pollutant buildup (mass). B and W have the same mass units.

Rating Curve wash-off assumes that the rate of wash-off W (mass/time) is proportional to the runoff rate Q raised to the power of C2w (James et al. 2010, Rosmann 2015):

for06     (6)

Street sweeping is one of the most effective ways of controlling and reducing washed-off pollution from surfaces in high traffic areas (James et al., 2010). The efficiency of street sweeping regarding the pollution removal depends on many factors, including the type of street sweeping equipment, sweeping frequency, the presence of parked cars, the amount of total solid matter, the pollutant being considered and the frequency of rain events. Existing studies suggest that street sweeping is more effective in removing constituents that are bound to solid particles (James et al., 2010).

This paper illustrates coupled modelling of quantity and quality of runoff generated from a bridge deck of Gazela highway bridge. The Gazela Bridge is the busiest Belgrade bridge spanning the Sava River, with over 150,000 vehicles passing over it daily. The study has two main objectives. First, to assess runoff flow rates and compare them to the receiving capacity of the public sewers where runoff is discharged, and second, to assess runoff quality and to define street sweeping strategies for reducing bridge runoff pollution.

 

Runoff and Pollution Modelling

Description of the Gazela Bridge

The Gazela Bridge is a highway bridge spanning the Sava River in Belgrade (Serbia) built in the early 1970s. A general reconstruction of the bridge was completed in 2012, during which a closed drainage system for roadway runoff and other catchment areas on the bridge and access roads were installed. The bridge storm water drainage system is divided into two parts – catchments: the left (west) and the right (east) sections of the bridge, each discharging collected runoff into the public sewers located along the river banks on both sides of the river. This paper analyzes west section of the Gazela Bridge, consisting of a steel-concrete section over the river, a concrete section on the left river bank and two reinforced concrete ramps connecting the highway to Vladimira Popovića Street. The longitudinal slope of the road is 1.3% on the steel/concrete part of the bridge, while the slope of the concrete section of the bridge and the access ramps is between 1.5% and 6%. The lateral slope of the road is about 2%.

The analyzed west section of the bridge is made up of impermeable - asphalt surfaces with a total area of about 13.600 m2. Storm runoff is collected by 27 curb gullies constructed along both sides of the steel-concrete section of the bridge and by another 8 gullies on the concrete section of the bridge. The distance between the gullies is 11.25 m on the steel part of the bridge and from 8 to 9 m on the concrete part of the bridge. On the descending/ascending ramps of the bridge rainwater runoff is collected by 38 curb gullies constructed along the right edge of the traffic lanes. The average distance between the gullies is from 9 to 11 m. A schematic view of the considered west section of the Gazela Bridge is shown in Figure 2.

Storm runoff collected by the gullies is discharged into the collector pipelines, with a diameter of Ø300-Ø600 mm, which are suspended on the bridge structure. The collected runoff is transported to a manhole under the bridge which serves as a divider node. From this manhole main collector pipe continues towards the discharge point, and from where one collector pipe continues, with lateral branch pipes are feeding an underground system for partial treatment, retention and infiltration. Collected storm water from the bridge deck is eventually discharged into the public sewer network of the Public Utility Company Belgrade Waterworks and Sewerage (PUC BWS). The reason for the construction of a stormwater management system for treatment, retention and infiltration of rainwater is for limiting rainwater flow volumes being discharged into the public storm sewer network to 140 l/s, as required by the sewer system operator - PUC BWS.

 

fig02
Figure 2: Modeled part of the Gazela Bridge.

 

Rainfall – Runoff Modeling

The rainfall – runoff modeling was performed in the software provided by the United States Environmental Protection Agency (EPA). EPA's Storm Water Management Model (SWMM) is freely available (Rossman, 2015) and globally recognized as most prominent software for hydraulic analysis of sewage networks. SWMM hydraulic model, set up for the purpose of this study, included impervious surfaces on the west side of Gazela Bridge and access roads, totaling at 13600 m2 (Figure 3). The catchment area for each gully was determined based on the 3D model of the bridge surface, acquired from the Gazela Bridge Reconstruction Design. The model included 146 impermeable drainage surfaces, each draining into a corresponding gully (Figure 3). Each gully is connected to a piped sewage system that further channels the runoff into the collecting shaft K3, and re-directed flow into the public sewer system.

Flow routing in drainage system is modeled using a dynamic wave model. It has been assumed that the runoff starts when the accumulation of water on the surface exceeds 0.5 mm. This paper will focus on the analyses of the bridge drainage system only, without storm water management system for treatment, retention and infiltration of collected runoff.

 

fig03
Figure 3: Model of the west section of the Gazela bridge drainage system (SWMM).

 

Rainfall Data

Rainfall runoff modeling was performed for two different types of rain:

  • uniform rainfall,
  • observed rainfall data in the period from March 24 to August 23, 2014 measured at the rain gauge located on the premises of the Faculty of Civil Engineering in Belgrade (GRF). The rain gauge operates on a tipping bucket principle with a 0.2 mm measurement step. An integration time step of 5 min was adopted for calculation of the cumulative rainfall curve (Figure 4).

 

fig04
Figure 4: Cumulative rainfall for the rain gauge "GRF", integration time step Δt = 5 min.

 

Pollution Modeling – Build-up and Wash-off

Modeling of pollution build-up and wash-off was done for the following parameters: COD - Chemical oxygen demand; TSS - Total Suspended Solids; TN- total nitrogen; TP- total phosphorus; Fe- iron; Zn- zinc; Cr- chrome; Cu- copper.

An overview of pollutant build-up and wash-off functions and coefficients used for modeling is given in Table 1. Study results on pollution accumulation on asphalt surfaces that were conducted at the Faculty of Civil Engineering in Belgrade (Djukić 2014, Djukić et al., 2018) were used for build-up modeling. For pollutant wash-off functions and coefficients are adopted from literature (Wicke et al. 2012).

 

tab01

 

Results and Discussion

Pollution Build-up and Wash-off

In the first step, pollution build-up and wash-off are analyzed for three 10 minute uniform rainfall events with total precipitation depths: 5 mm, 10 mm and 20 mm. The calculations were carried out for 2, 5 and 10 antecedent dry days. Figure 5 shows the total amount (W) of washed-off TSS and Zn for various rainfall depths and antecedent dry days. From the results it can be observed that the amount of wash-off pollutants increases with rainfall depth, but that the increase decreases with the increase of rainfall depth. It can be concluded that a 10 mm uniform rain washes off practically all pollution build-up.

 

fig05
Figure 5: Amount of washed-off off (W) TSS and Zn for various rainfall depths and antecedent dry days.

 

Strom Runoff Flow Rates

Figure 5 shows the storm runoff hydrograph from the entire analyzed catchment (west bridge section) for the observed rainfall in the period 24 March - 23 August 2014 calculated using the dynamic wave model. It can be concluded that, within the 5-month analysis period, there are 4 rain episodes that generate an outflow greater than 140 l/s (Figure 6), which has been defined as the maximum acceptable storm runoff flow rate that can be discharged into the public storm sewers on the west (left) river bank.

 

fig06
Figure 6: Hydrograph of runoff for the west side of the Gazela Bridge for the rainfall observed in the period March – August 2014.

 

Storm Water Quality

Results of storm runoff quality reveal that the maximum runoff pollution concentrations reach 200-1000 mg/L for COD and 1000-5000 mg/L for TSS. Such values are higher than those occurring in municipal wastewater, but the duration of these extreme values is very short. What is more important regarding the pollution impact is the total amount of washed-off pollutants over a certain time period. Figure 7 shows the model results for daily quantities of washed-off TSS, which may be regarded as the primary pollutant present in the stormwater road runoff. Daily amounts of washed-off TSS that can go as high as 180 kg/day, or 2600 p.e. (p.e. – population equivalent, where 1 p.e. is equal to 70 gTSS/day, as per ATV-DVWK-A 131E standard).

 

fig07
Figure 7: Daily quantities of TSS from the basin on the west section of the Gazela Bridge for the observed rainfall in the period March – August 2014.

 

The Impact of Street Sweeping on Pollutant Wash-off

The impact of the street sweeping on the total amount of pollutant wash-off was analyzed by the rainfall-runoff model of the bridge deck described above, coupled with pollutant build-up and wash-off models, for the observed rainfall data collected from the GRF rain gauge. A dry street sweeping (cleaning) method was adopted, and it was assumed that street sweeping is performed at regular time intervals. The calculations were made for different intervals between two consecutive street sweeping: cleaning twice a day, once a day, every 2, 7 and 14 days. The initial conditions for each model were the same and included pollutants accumulated on road surfaces during a total of 10 antecedent dry days. Values of 30% and 60% were adopted for percentages of pollutants that can be removed from the roadway in one pass of the road sweeping equipment (sweeping efficiency). It was assumed that a total of up to 80% of the pollutants analyzed would be removed from the road surfaces by sweeping.

Figure 8 shows the model results for the total washed-off TSS, over analyzed period of 5 months, for the various sweeping frequencies and removal efficiencies. The dashed horizontal line indicates the total washed-off TSS value from the analyzed part of the bridge with no sweeping applied. Figure 8 clearly shows that the amount of washed-off pollution increases rapidly as the time interval between two consecutive road sweepings increases.

 

fig08
Figure 8: Calculated total washed-off TSS over the period from 24 March to 23 August 2014 for various time intervals between consecutive sweepings and for sweeping efficiencies of 30% and 60%.

 

Conclusions

The results of the analysis confirm that the amount of pollution wash-off increases with increasing rainfall, but the wash-off increase rate gradually decrease with the increase of rainfall depth.. The results indicate that a uniform rainfall of approximate 10 mm washes off practically all accumulated pollutants from analyzed bridge deck surface. Analyzing the model results of rainfall-runoff analyses for the observed rainfall in the period March 23 - August 23, 2014, measured by the rain gauge at the Faculty of Civil Engineering in Belgrade, it can be concluded that during the analyzed period there were 4 rain events that generated runoff that was higher than the maximum flow that can be accepted into the public sewer system. This speaks in favor of the justifying the application of methods for reducing peak flows by means of temporary runoff retention prior its discharge into the public storm sewer system.

Model indicates that pollution concentrations in the runoff can reach higher values than those occurring in communal wastewater, but the duration of high concentrations is very short. The most pronounced form of pollution present in the bridge runoff is TSS, and the daily quantities of TSS in the runoff is estimated to be as high as 180 kg/day, or about 2600 p.e.

The conducted analyses lead to conclusion that road sweeping can significantly reduce pollution wash-off. However, model results indicate that in order to reduce total amounts of washed-off pollutants by 50% or more, sweeping frequency should be high – approximately twice per day. Also, results reveal that by increasing the time interval between two consecutive road sweepings the amount of total washed off pollutants rapidly increase. It is particularly emphasized here that the street sweeping method discussed herein involves the removal of accumulated materials from road surfaces by various methods which also include disposal of removed material in a place where they will not endanger water resources and the environment. For the above reasons, the road cleaning method which include washing of surfaces with high amounts of water (a method commonly used in Serbia), is not an adequate from environmental protection point of view, as such method actually increases wash-off of accumulated pollution from the road surfaces into the environment.

 

Acknowledgement

The results published in this paper were realized within the project of technological development of the Ministry of Education, Science and Technological Development of the Republic of Serbia (Project TR 37010 and TR37009).

 

References

Bertrand-Krajewski, J.-L. 2007. Stormwater pollutant loads modeling: Epistemological aspects and case studies on the influence of field data sets on calibration and verification. Water Science and Technology 55 (4): 1–17.

Djukić A.(2016), Urban Storm Runoff Pollution Emission Modelling, PhD thesis (in Serbian), University of Belgrade, Faculty of Civil Engineering

Djukić A., Lekić B., Rajaković-Ognjanović V., Naunovic Z., Prodanović D. (2018) Build-up and characterisation of pollutants on urban impervious surfaces, Water Science and Technology 77 (8), 2123-2133

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

James W., Rossman L.A., James W.R.C. (2010) User's guide to SWMM 5, 13th Edition, Published by CHI, Guelph, Ontario, Canada. ISBN 978-0-9808853-5-4

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

Sabin L.D., Lim J.H., Venezia M.T., Winer A.M.,. Schiff K.C., Stolzenbach K.D. (2006) Dry deposition and resuspension of particle-associated metals near a freeway in Los Angeles. Atmospheric Environment 40 (2006) 7528–7538

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