Abstract

Stormwater runoffs are one of the primary causes for deteriorating water quality in the Nainital Lake, India – a prominent tourist attraction and the sole drinking water source for the habitants of Nainital City. Treatment of fluctuating runoffs and contaminant influxes before mixing with the lake's water by conventional methods would require a large land footprint, which is a big constraint in the Nainital because of the hilly region. Ballasted sand flocculation (BSF) technology requires much less land footprint; a full-scale 1 MLD capacity pilot plant was applied for treatment of stormwater runoffs of the Nainital Lake. Twenty-eight storm events were monitored for runoff characterization and for evaluating the performance of BSF technology. The runoff water showed marked variation especially for total suspended solids (TSS), total phosphorus (TP), chemical oxygen demand (COD), biochemical oxygen demand (BOD), total coliform (TC) and fecal coliform (FC) with maximum concentrations of 964 mg/l, 2.35 mg/l, 520 mg/l, 299 mg/l, 21 × 105 MPN/100 ml and 14 × 104 MPN/100 ml. The performance analyses results of the pilot plant revealed that the contaminants including trace metals in the stormwater runoff were reduced appreciably and the pollutant removal efficiencies were found to be largely unaffected by fluctuation of the influent contaminants' concentration.

INTRODUCTION

Stormwater runoffs from urban areas contain a wide variety of pollutants from both anthropogenic and natural sources (Perdikaki & Mason 1999; Kim & Kang 2004) and act as a major contributor to the pollution of receiving water bodies (Lee et al. 2004). Surface water bodies such as lakes face the problem of siltation and eutrophication due to the catchment's stormwater runoff (Landon et al. 2006). Effective stormwater management is of crucial importance for safeguarding surface-water bodies, especially those which are used for water supply, from the risk of contaminants from stormwater runoffs.

Nainital Lake, a lake of national importance in India, is one such water body that faces deteriorating water quality because of intervening activities in the catchment of the lake (NIH 2000; Dash et al. 2008). The Nainital Lake is the main source of water supply for Nainital City, a popular hill station situated around 1,937 m above mean sea level (MSL), in the foothills of the outer Himalayas. The lake also attracts thousands of tourists every year due to its scenic beauty. The economy of Nainital region is directly or indirectly dependent on the lake. The city, except for some slum areas, is covered by a sewerage system. Most of the slum residents have their own latrine in their houses with soak pit or septic tank facilities. A few families do not have a latrine at home and therefore they use either a community latrine or open defecation. A number of stormwater drainage channels located in the lake catchment area carry polluted urban stormwater runoff into the lake. Numerous earlier studies (Choudhary et al. 2009) have evaluated the physical, chemical and biological water characteristics of the Nainital Lake with no focus on characterization of stormwater quality and management of this seasonal influx, laden with pollutants.

There are many low impact development (LID)/best management practices (BMP) to remove water pollutants and/or reduce stormwater runoff peak and volume of runoff. Examples of LID/BMP practices include stormwater filtering system, infiltration system, detention basin, retention basin and constructed wetland etc. Stormwater filtering systems are typically designed solely for pollutant removal and generally applied to small drainage areas (less than 5 acres) and they are not typically cost-effective for the larger drainage areas (Claytor & Schueler 1996). BMPs, particularly ponds and wetlands, have higher or more reliable removal rates in comparison with other LID or BMP for nutrients, bacteria and hydrocarbons. In India, some researchers (Mittal et al. 2006; Vipat et al. 2008) have worked upon constructed wetlands for stormwater runoff treatment for the Delhi and Bhopal regions. These treatment techniques are usually not preferred where space is a constraint. Land availability is a critical factor in Nainital for the adoption of any treatment technology.

One of the technologies proven to be effective for surface water, combined sewer overflow (CSO) and primary/tertiary wastewater treatment is the ballasted sand flocculation (BSF), also known as high rate clarifier, which requires less space (EPA 2003; Kumar et al. 2016). The BSF technology is mainly based on a physical–chemical treatment process that uses a continuous recycled medium (sand) along with chemicals (coagulant and flocculant) to improve the settling properties of suspended solids through improved floc bridging (Plum et al. 1998; Landon et al. 2006; Gasperi et al. 2012). Hardly any attempt was made in India for stormwater runoff treatment at full scale or laboratory scale using BSF technology. In the present study, stormwater runoff was characterized and the performance of BSF was evaluated by employing it for treatment of stormwater runoffs of the Nainital Lake.

MATERIALS AND METHODS

Study area

Nainital Lake (Latitude 29° 23.127′ N and Longitude 79° 27.656′ E) has a crescent shape and is situated in the Nainital district of Uttarakhand, India. The maximum length, width and depth of the lake are 1.4 km, 0.45 km and 27.32 m, respectively, with a mean depth of 18.5 m (NIH 2000). The catchment area of the lake is 4.9 sq km of which 48.4% is covered by forest, 18.3% is barren, 19.3% is human settlement and 10.4% is water bodies. The average annual rainfall in the catchment is 2,030 mm. Naina Devi drain, a major drain in the lake's catchment that contributes 60% of the runoff to the Nainital Lake, has been selected for stormwater runoff characterization and performance evaluation of BSF technology. The catchment area of the drain is 2.36 sq km. Google Earth and pictorial views of Naina Devi drain are shown in Figure 1(a) and 1(b). A pilot plant (BSF system) as shown in Figure 1(c) and 1(d) was installed near the Naina Devi drain to tap 1 MLD (42 m3/hr) stormwater from the drain.

Figure 1

Location map: (a) Google Earth view of the Nainital Lake and Naina Devi drain; (b) pictorial view of the drain; (c) schematic of the BSF system; (d) BSF pilot plant at study site.

Figure 1

Location map: (a) Google Earth view of the Nainital Lake and Naina Devi drain; (b) pictorial view of the drain; (c) schematic of the BSF system; (d) BSF pilot plant at study site.

Description and operation of BSF system

The experiments were conducted on a pilot-scale BSF system (Figure 1(c) and 1(d)) for treatment of stormwater runoff produced from storm events during the monsoon seasons of 2016 and 2017. The BSF system was comprised of three processes; injection, maturation and settling. Prior to entering into the first tank (coagulation tank), influent was passed through a bar screen for screening of floating substances. In the first process, coagulant is added to the coagulation tank and flocculant with micro-sand is injected into the flocculation tank. The second process is maturation in which the ballast material (micro-sand) enhances floc formation, which results in a faster settling rate. The third stage of the process is settling, where the floc settles by gravity through inclined plate settlers, which further enhance the settling process by providing a greater surface area. Sludge containing ballast is collected in a cone-shaped chamber. Ballast from the bottom of the chamber is separated from the sludge through a hydro-cyclone and reintroduced into the flocculation tank.

The technical design of the plant is in Table 1. The total hydraulic retention time (HRT) of the plant was 16 min, which corresponds to a 2-2-6-6 design, which means 2 min retention time in the coagulation tank, 2 min in the flocculation tank, 6 min in the maturation tank and 6 min settling time. The coagulation, flocculation and maturation tanks were equipped with mixers, which provide high mixing speed (160–180 rpm) in the first two tanks and low mixing speed in the maturation tank (40–60 rpm).

Table 1

Technical details of BSF pilot plant

ParametersUnitValue
Design flow m3/hr 42 
Coagulation tank size m3 1.62 
Flocculation tank size m3 1.62 
Maturation tank size m3 5.76 
Land footprint m2/m3/hr 1.29 
HRT minutes 16 
Rise rate m/hr 40 
Energy consumption kWh/m3 0.35 
ParametersUnitValue
Design flow m3/hr 42 
Coagulation tank size m3 1.62 
Flocculation tank size m3 1.62 
Maturation tank size m3 5.76 
Land footprint m2/m3/hr 1.29 
HRT minutes 16 
Rise rate m/hr 40 
Energy consumption kWh/m3 0.35 

The chemical dosages of the plant were optimized by conducting modified jar tests in the laboratory before the field application of the plant. Modified jar tests were conducted following the procedures of Ding et al. (1999) and Zhu et al. (2007) in transparent 1 L beakers placed on a standard jar test apparatus equipped with an impeller. Alum (Al2(SO4)3.18H2O) as coagulants, cationic polymer as flocculant and micro-sand (silica) of size 130 μm as ballast were used for the study. Jar tests were conducted for different rainfall intensities corresponding to turbidity values of 52, 92, 204, 572 and 698 NTU. For each jar test, coagulant and flocculant dose corresponding to minimum residual turbidity were chosen as optimum doses. The optimum doses for the first, second, third, fourth and fifth jar tests were found to be 20, 25, 40, 70 and 80 mg/l for coagulant and 0.5, 0.5, 0.7, 1.0, 1.2 mg/l for flocculant dose. Hence, the optimal dosage for observed turbidity during the study period was 20–80 mg/l for alum as coagulant and 0.5–1.2 mg/l for cationic flocculant. Many researchers (Ding et al. 1999; Jolis & Ahmad 2004; Cornwell et al. 2010) have reported that the performance of the BSF process is effective when sand sizes and dose are in the range of 75–250 μm and 10–12 g/l for stormwater/surface water treatment, therefore in the present study 130 μm micro-sand was considered and 10 gm/l was the dose was applied. Caustic soda (NaOH) was not added for pH correction as alkalinity was high in the influent.

Sample collection and analyses

Data of 28 storm events were used for the performance evaluation of the BSF system. The plant was operated for 2–4 hours for each storm event depending upon the rainfall and runoff. The characteristics of storm events are presented in Table 2. Water samples from inlet and outlet (total no. of samples – 214) of the BSF system were collected for all storm events. Numbers of samples for each storm event varied from three to eight depending upon the duration of the event. Samples were collected after 10 minutes of rainfall initiation and thereafter at every 20-minute interval until the flow in the drain reached its original flow condition. The samples were analyzed for pH, alkalinity, turbidity, total suspended solids (TSS), chemical oxygen demand (COD), biochemical oxygen demand (BOD), ammoniacal nitrogen (NH4+-N), nitrate nitrogen (NO3-N), ortho and total phosphorus (OP and TP). Bacteriological parameters of total coliform (TC) and fecal coliform (FC) and heavy metals (Cu, Zn, Cd, Cr, Pb) were also analyzed. The samples were analyzed as per the standard methods (APHA 2012).

Table 2

Characteristics of monitored storm events in the study area

DateStorm event no.Intensity (mm/hr)Peak runoff (m3/s)DateStorm event no.Intensity (mm/hr)Peak runoff (m3/s)
22-08-2016 30.0 6.66 05-07-2017 15 49.5 7.80 
26-08-2016 12.0 3.10 06-07-2017 16 8.6 2.60 
28-08-2016 6.9 1.72 07-07-2017 17 7.0 1.92 
08-09-2016 7.2 1.75 10-07-2017 18 18.0 4.90 
11-09-2016 6.0 1.60 21-07-2017 19 2.4 0.90 
12-09-2016 3.6 1.28 24-07-2017 20 3.0 1.15 
18-09-2016 7.5 1.90 25-07-2017 21 18.0 4.60 
24-09-2016 9.2 2.60 28-07-2017 22 4.0 1.55 
25-09-2016 7.5 2.10 30-07-2017 23 16.6 4.30 
16-06-2017 10 4.0 1.52 02-08-2017 24 19.2 5.10 
18-06-2017 11 24.0 5.98 09-08-2017 25 3.2 1.22 
01-07-2017 12 15.8 4.50 15-08-2017 26 6.9 1.70 
02-07-2017 13 2.4 1.10 16-08-2017 27 2.7 1.12 
04-07-2017 14 8.0 2.30 18-08-2017 28 4.8 1.75 
DateStorm event no.Intensity (mm/hr)Peak runoff (m3/s)DateStorm event no.Intensity (mm/hr)Peak runoff (m3/s)
22-08-2016 30.0 6.66 05-07-2017 15 49.5 7.80 
26-08-2016 12.0 3.10 06-07-2017 16 8.6 2.60 
28-08-2016 6.9 1.72 07-07-2017 17 7.0 1.92 
08-09-2016 7.2 1.75 10-07-2017 18 18.0 4.90 
11-09-2016 6.0 1.60 21-07-2017 19 2.4 0.90 
12-09-2016 3.6 1.28 24-07-2017 20 3.0 1.15 
18-09-2016 7.5 1.90 25-07-2017 21 18.0 4.60 
24-09-2016 9.2 2.60 28-07-2017 22 4.0 1.55 
25-09-2016 7.5 2.10 30-07-2017 23 16.6 4.30 
16-06-2017 10 4.0 1.52 02-08-2017 24 19.2 5.10 
18-06-2017 11 24.0 5.98 09-08-2017 25 3.2 1.22 
01-07-2017 12 15.8 4.50 15-08-2017 26 6.9 1.70 
02-07-2017 13 2.4 1.10 16-08-2017 27 2.7 1.12 
04-07-2017 14 8.0 2.30 18-08-2017 28 4.8 1.75 

RESULTS AND DISCUSSION

Table 3 summarizes the physico-chemical and microbiological characteristics of influent and effluent of the BSF system and removal rate of pollutants monitored during the study period. The pollutants detected and the corresponding average concentration ranges are comparable with data reported in the literature for stormwater runoff (Joshi & Al Obaidy 2014). The values of TSS, TP, COD, BOD, TC and FC reached up to 964 mg/l, 2.35 mg/l, 520 mg/l, 299 mg/l, 21 × 105 MPN/100 ml and 14 × 104 MPN/100 ml respectively (Table 3), which may be due to mixing of sewage because of bursting or overflow from the sewer pipeline. Total phosphorus concentration was in the range 0.1–2.35 mg/l, which is much higher than the threshold value (20 μg/l) causing eutrophication in the lake.

Table 3

Characteristic of influent, effluent and percentage removal of pollutants

ParametersInfluent concentrationEffluent concentration% Removal
average ± SD (min–max)average ± SD (min–max)average ± SD (min–max)
pH 7.45 ± 0.28 (6.80–8.20) 7.25 ± 0.28 (6.6–8) – 
Alkalinity (mg/L as CaCO3247 ± 66 (80–380) 202 ± 63 (40–338) – 
Turbidity (NTU) 171 ± 165 (12–724) 2.55 ± 0.80 (1.4–6.0) 96 ± 3 (85–99) 
TSS (mg/L) 216 ± 176 (38–964) 15 ± 11 (4–52) 91 ± 6 (64–98) 
COD (mg/L) 158 ± 121 (21–520) 38 ± 21 (9–125) 72 ± 10 (43–88) 
BOD (mg/L) 75 ± 64 (7–299) 19 ± 13 (3–66) 70 ± 12 (29–82) 
NH4+-N (mg/L) 2.42 ± 1.60 (0.15–7.20) 2.14 ± 1.45 (0.12–6.50) 10 ± 29 (0–31) 
NO3-N (mg/L) 1.12 ± 0.84 (0.10–3.70) 1.16 ± 0.89 (0.08–3.90) No removal 
PO4-P (mg/L) 0.36 ± 0.26 (0.05–1.40) 0.04 ± 0.02 (0.01–0.06) 86 ± 8 (60–95) 
TP (mg/L) 0.66 ± 0.38 (0.1–2.35) 0.07 ± 0.03 (0.02–0.12) 88 ± 6 (63–96) 
Total coliform(MPN/100 ml) 45 × 104 ± 68 × 104 (43 × 102–21 × 10515 × 103 ± 27 × 103 (43 × 101–93 × 10386 ± 6 (78–95) 
Fecal coliform (MPN/100 ml) 14 × 103 ± 27 × 103(15 × 101–14 × 10459 × 101 ± 80 × 101 (2 × 101–39 × 10284 ± 5 (61–96) 
Cu (μg/L) 38 ± 22 (5–73) 12 ± 7.0 (3–32) 65 ± 11 (40–82) 
Zn (μg/L) 625 ± 137 (368–890) 226 ± 56 (126–327) 64 ± 7 (51–77) 
Cd (μg/L) 1.60 ± 1.00 (0.5–4.8) 0.49 ± 0.40 (0.1–1.9) 63 ± 13 (33–88) 
Cr (μg/L) 45 ± 31 (5–127) 19 ± 14 (2–66) 56 ± 10 (33–78) 
Pb (μg/L) 25 ± 19 (7–83) 8 ± 6 (2–32) 67 ± 8 (56–87) 
ParametersInfluent concentrationEffluent concentration% Removal
average ± SD (min–max)average ± SD (min–max)average ± SD (min–max)
pH 7.45 ± 0.28 (6.80–8.20) 7.25 ± 0.28 (6.6–8) – 
Alkalinity (mg/L as CaCO3247 ± 66 (80–380) 202 ± 63 (40–338) – 
Turbidity (NTU) 171 ± 165 (12–724) 2.55 ± 0.80 (1.4–6.0) 96 ± 3 (85–99) 
TSS (mg/L) 216 ± 176 (38–964) 15 ± 11 (4–52) 91 ± 6 (64–98) 
COD (mg/L) 158 ± 121 (21–520) 38 ± 21 (9–125) 72 ± 10 (43–88) 
BOD (mg/L) 75 ± 64 (7–299) 19 ± 13 (3–66) 70 ± 12 (29–82) 
NH4+-N (mg/L) 2.42 ± 1.60 (0.15–7.20) 2.14 ± 1.45 (0.12–6.50) 10 ± 29 (0–31) 
NO3-N (mg/L) 1.12 ± 0.84 (0.10–3.70) 1.16 ± 0.89 (0.08–3.90) No removal 
PO4-P (mg/L) 0.36 ± 0.26 (0.05–1.40) 0.04 ± 0.02 (0.01–0.06) 86 ± 8 (60–95) 
TP (mg/L) 0.66 ± 0.38 (0.1–2.35) 0.07 ± 0.03 (0.02–0.12) 88 ± 6 (63–96) 
Total coliform(MPN/100 ml) 45 × 104 ± 68 × 104 (43 × 102–21 × 10515 × 103 ± 27 × 103 (43 × 101–93 × 10386 ± 6 (78–95) 
Fecal coliform (MPN/100 ml) 14 × 103 ± 27 × 103(15 × 101–14 × 10459 × 101 ± 80 × 101 (2 × 101–39 × 10284 ± 5 (61–96) 
Cu (μg/L) 38 ± 22 (5–73) 12 ± 7.0 (3–32) 65 ± 11 (40–82) 
Zn (μg/L) 625 ± 137 (368–890) 226 ± 56 (126–327) 64 ± 7 (51–77) 
Cd (μg/L) 1.60 ± 1.00 (0.5–4.8) 0.49 ± 0.40 (0.1–1.9) 63 ± 13 (33–88) 
Cr (μg/L) 45 ± 31 (5–127) 19 ± 14 (2–66) 56 ± 10 (33–78) 
Pb (μg/L) 25 ± 19 (7–83) 8 ± 6 (2–32) 67 ± 8 (56–87) 

The BSF system is highly effective in removing particulates/solids as evident from turbidity (96%) and TSS (91%) removal. COD and BOD removal efficiencies that vary between 43% and 88% for COD and between 29% and 82% for BOD indicate that some colloidal particles can be coagulated and removed. Phosphorus removal rate ranged between 60% and 95% for OP and between 63% and 96% for TP. Phosphorus removal is important as it is the limiting factor for eutrophication of the Nainital Lake (NIH 2000). It is hypothesized that the mechanism for phosphorus removal is precipitation and adsorption. The species Al(OH)3 may be responsible for phosphorus precipitation, and AlOH2+ and Al(OH)2+ are responsible for adsorption (Yang et al. 2010). The event-wise removal of TSS, COD, OP and TP are shown in Figure 2(a) and 2(b) respectively. It is noticed that the deviation of removal efficiencies from the mean value generally lies within 10% but there is large fluctuation in influent concentration. The removal rate of ammoniacal nitrogen found varied from 0% to 31% with no removal of nitrate nitrogen. The removal efficiency of pollutants is comparable with that reported in other studies based on the performance of BSF systems for CSO/surface water of similar characteristics of influent (Plum et al. 1998; Gasperi et al. 2012).

Figure 2

Average inlet, outlet contaminant concentration and removal efficiency during storm events for (a) TSS and COD, (b) OP and TP.

Figure 2

Average inlet, outlet contaminant concentration and removal efficiency during storm events for (a) TSS and COD, (b) OP and TP.

Higher coliform concentration may cause management concerns in terms of public and ecological health and can affect the ability of water bodies to achieve their intended uses. The inlet concentrations of TC and FC were found to be in the range 3–6 log units and 2–4 log units, respectively. The average removals of TC and FC in the BSF system were found to be 1.5 log and 1.2 log respectively. The result matches with the World Health Organization (WHO 2004) report stating that coagulation, flocculation and sedimentation can result in 1–2 log removals of bacteria. With respect to coagulation and flocculation, most bacteria may be considered as colloidal or suspended particles. Removal of coliforms may be due to destabilization of particles by neutralizing or reducing their surface electrical charge, enmeshing or bridging them in floc particles.

Heavy metals such as, Cu, Cd, Cr, Pb, and Zn in urban stormwater can degrade the ecological balance of the receiving waters and can be harmful for plants and humans due to the toxicity levels and degradation resistance. Since heavy metals do not degrade, their removal from stormwater runoff has been a concern in recent years. In this study, the intent for the metal analyses was to know the influx of trace metals in the lake and evaluate the potential of BSF to remove various metal constituents. The results showed anthropogenic enrichment of Pb, Cr and Zn in the runoff water. The larger concentrations of these metals are due to mixing of municipal sewage, excessive vehicular movement, uses of paints particularly on tin roofs, automobile repair and welding shops, atmospheric deposition and soil erosion, etc. Heavy metals in urban runoff are reported to be found in particulate form or bound to particulates and tend to settle out (Muthukrishnan 2006). In the present study also, it is found that trace metals (Cd, Cr, Cu, Zn and Pb) show a positive correlation with TSS, as illustrated in Figure 3 for Zn and Pb. It may be interpreted from Figure 3 that a larger fraction of metals may be present in particulate form and therefore it is expected that metal removal occurs by sedimentation of particle-associated metals.

Figure 3

Scatter plot between Zn/Pb with TSS.

Figure 3

Scatter plot between Zn/Pb with TSS.

Some studies (El Samrani et al. 2008) suggested that dissolved metal elimination in the soluble fraction is due to their sorption by iron or aluminium hydrolysis products. The removal efficiency of Cu, Zn, Cd, Cr and Pb for the present case was found to vary as 40%–82%, 51%–77%, 33%–88% 33%–78% and 56%–87%, respectively (Figure 4).

Figure 4

Removal efficiencies of the BSF system for heavy metals.

Figure 4

Removal efficiencies of the BSF system for heavy metals.

Based on the removal efficiencies, the pollutants are classified in three categories: (i) efficient removal when the removal rate is greater than 80%, (ii) moderate removal for removal rate >50% and ≤80%, (iii) weak removal for a rate up to 50%. Efficiently removed pollutants are turbidity, TSS, OP, TP, TC and FC; moderately removed pollutants are COD, BOD, Zn, Cu, Cr, Cd and Pb; and weakly removed pollutants are NH4+-N and NO3-N.

The efficiency of treatment technologies depends on the operational variables and hence, the comparisons based on published literature are grim. However, an attempt was made to compare the removal efficiency observed in this study with the published data for other treatment techniques, namely, filter strip, sand filter, infiltration system, detention basin, retention basin and constructed wetland (Table 4), generally used for stormwater runoff treatment. The removal efficiency of the BSF system for TSS, TP and FC was better than the other techniques, whereas for removal of metals, the results were comparable. Moreover, the studies by Yu et al. (1993) and Lucas et al. (2015) indicated an increase in concentration of certain parameters occasionally in the treated effluents (percentage removal value is negative). The increase in the concentration of these parameters was due to re-suspension of particles for the filter strip and constructed wetland techniques. However, such cases were not observed in the BSF system. These eventually suggest that the performance of BSF technology that requires less land space is better than the other technologies in removal of pollutants.

Table 4

Pollutant removal efficiency (%) of different treatment techniques

Treatment techniquesTSSTPBODFCZnPbCuReferences
Filter strip 70 10–40 – 40–80 Metals: 40–50 Claytor & Schueler (1996)  
54–84 (−)25–40 – – 47–55 – – Yu et al. (1993)  
Sand filter 89 59 – 65 Metals: 72–86 Glick et al. (1998)  
50–80 50–80 – <30 Metals: 50–80 US EPA (1993)  
Infiltration system 50–80 50–80 – 65–100 Metals: 50–80 US EPA (1993)  
Dry detention basin 70 13 – – 57 62 – Weiss et al. (2005)  
Retention basin 50–80 30–65 – <30 Metals: 50–80 US EPA (1993)  
54 46 39 46 69 76 – Weiss et al. (2005)  
Constructed wetland 50–80 15–45 – <30 Metals: 50–80 US EPA (1993)  
58 45 – – 29 74 65 Carleton et al. (2000)  
46 12 – – 52 65 65 Birch et al. (2004)  
(−)16–99 (−)18–48 (−)63–99 78–98 (−)56–99 0–98 (−)97–97 Lucas et al. (2015)  
BSF 64–98 63–96 29–82 61–96 51–77 56–87 40–82 Present study 
Treatment techniquesTSSTPBODFCZnPbCuReferences
Filter strip 70 10–40 – 40–80 Metals: 40–50 Claytor & Schueler (1996)  
54–84 (−)25–40 – – 47–55 – – Yu et al. (1993)  
Sand filter 89 59 – 65 Metals: 72–86 Glick et al. (1998)  
50–80 50–80 – <30 Metals: 50–80 US EPA (1993)  
Infiltration system 50–80 50–80 – 65–100 Metals: 50–80 US EPA (1993)  
Dry detention basin 70 13 – – 57 62 – Weiss et al. (2005)  
Retention basin 50–80 30–65 – <30 Metals: 50–80 US EPA (1993)  
54 46 39 46 69 76 – Weiss et al. (2005)  
Constructed wetland 50–80 15–45 – <30 Metals: 50–80 US EPA (1993)  
58 45 – – 29 74 65 Carleton et al. (2000)  
46 12 – – 52 65 65 Birch et al. (2004)  
(−)16–99 (−)18–48 (−)63–99 78–98 (−)56–99 0–98 (−)97–97 Lucas et al. (2015)  
BSF 64–98 63–96 29–82 61–96 51–77 56–87 40–82 Present study 

CONCLUSION

The following conclusions are drawn from the study:

  • The stormwater runoff flowing through Naina Devi drain contains high concentrations of total suspended solids, organics (COD and BOD), nutrient (TP), heavy metals (Pb, Cr, Zn) and pathogenic indicators (TC and FC), thus an effective treatment of stormwater runoff is required before it finds its way into the lake.

  • The BSF system was capable of effectively removing the particulate matter, organic constituent, nutrients (phosphorus) and coliforms from stormwater runoff with average removal rates of 96% for turbidity, 91% for TSS, 72% for COD, 70% for BOD, 86% for OP and 88% for TP. The log removal of TC and FC has been achieved as 1.5 and 1.2 respectively during the present study. The average removal rate of heavy metals, viz. Cu, Zn, Cd, Cr and Pb, were 65%, 64%, 63%, 56% and 67% respectively.

  • Performance comparison with other stormwater runoff treatment techniques showed that the BSF system is an efficient and promising technique for stormwater runoff treatment and for mitigating the adverse effects of urban stormwater runoffs on the Nainital Lake water environment.

ACKNOWLEDGEMENT

The European Commission within the 7th Framework Programme under grant number 308672 and the Department of Science and Technology, Government of India, are kindly acknowledged for their financial support in this study. The authors would also like to kindly acknowledge Uttarakhand PeyJal Nigam and AKTION INDIAA, Ahmedabad for their support.

REFERENCES

REFERENCES
APHA
2012
Standard Methods for the Examination of Water and Wastewater
, 22nd edn.
American Public Health Association/American Water Works Association/Water Environment Federation
,
Washington, DC
,
USA
.
Birch
G. F.
,
Matthai
C.
,
Fazeli
M. S.
&
Suh
J. Y.
2004
Efficiency of a constructed wetland in removing contaminants from stormwater
.
Wetlands
24
(
2
),
459
466
.
Carleton
J. N.
,
Grizzard
T. J.
,
Godrej
A. N.
,
Post
H. E.
,
Lampe
L.
&
Kenel
P. P
, .
2000
Performance of a constructed wetlands in treating urban stormwater runoff
.
Water Environment Research
72
(
3
),
295
304
Choudhary
P.
,
Routh
J.
,
Chakrapani
G. J.
&
Kumar
B.
2009
Biogeochemical records of paleoenvironmental changes in Nainital Lake, Kumaun Himalayas, India
.
J. Paleolimnol.
42
(
4
),
571
586
.
Claytor
R. A.
&
Schueler
T. R.
1996
Design of Stormwater Filtering Systems. Report prepared for Chesapeake Research Consortium, Inc. and US Environmental Protection Agency, Region 5. Center for Watershed Protection, Silver Spring, MD, USA.
Cornwell
D. A.
,
Tobiason
J.
&
Brown
R.
2010
Report on Innovative Application of Treatment Process for Spent Filter Backwash. Jointly Sponsored by Water Research Foundation and US EPA
.
Water Research Foundation
,
Denver, CO, USA
.
Dash
R. R.
,
Mehrotra
I.
,
Kumar
P.
&
Grischek
T.
2008
Lake bank filtration at Nainital, India: water-quality evaluation
.
Hydrogeology Journal
16
,
1089
1099
.
Ding
Y.
,
Dresnack
R.
&
Chan
P. C.
1999
Internal report on Assessment of High-Rate Sedimentation Processes: Micro-carrier Weighted Coagulation Jar Tests
.
Prepared for the US EPA, Contract Number 7C-R364-NAFX
,
NJ, USA
.
EPA
2003
Wastewater Technology Fact Sheet: Ballasted Flocculation. EPA 832-F-03-010 Office of Waste Management. Municipal Technology Branch, US Environmental Protection Agency, USA.
Glick
R.
,
Chang
G. C.
&
Barrett
M. E.
1998
Monitoring and evaluation of stormwater quality control basins
. In:
Watershed Management: Moving From Theory to Implementation
,
Water Environment Federation
,
Denver, CO, USA
, pp.
369
376
.
Kim
L. H.
&
Kang
J. H.
2004
Characteristics of first flush in highway storm runoff
.
J. Korean Society on Water Quality
20
(
6
),
641
646
.
Kumar
S.
,
Ghosh
N. C.
&
Kazmi
A. A.
2016
Ballasted sand flocculation for water, wastewater and CSO treatment
.
Environmental Technology Reviews
5
,
57
67
.
Landon
S.
,
Donahue
C.
,
Jeyanayagam
S.
&
Craden
D.
2006
Rain check: Columbus, Ohio, considers ballasted flocculation to treat its wet weather flows
.
Water Environment and Technology
18
,
30
35
.
Lee
H.
,
Lau
S. L.
,
Kayhanian
M.
&
Stenstrom
M. K.
2004
Seasonal first flush phenomenon of urban stormwater discharges
.
Water Research
38
(
19
),
4153
4163
.
Lucas
R.
,
Earl
E. R.
,
Babatunde
A. O.
&
Bockelmann-Evans
B. N.
2015
Constructed wetlands for stormwater management in the UK: a concise review
.
Civil Engineering and Environmental Systems
32
(
3
),
251
268
.
Mittal
A. K.
,
Jain
M.
,
Jamwal
P.
&
Mouchel
J. M.
2006
Treatment of urban run off using constructed wetlands in New Delhi, India
. In:
Proceedings of the World Environmental and Water Resource Congress 2006
(R. Graham, ed.),
American Society of Civil Engineers (ASCE)
,
Reston, VA, USA
.
Muthukrishnan
S.
2006
Treatment of heavy metals in stormwater runoff using wet pond and wetland mesocosms
. In:
Proceedings of the Annual International Conference on Soils, Sediments, Water and Energy
,
University of Massachusetts, Amherst, MA, USA
, Vol.
11
,
Article 9
, pp.
125
145
.
NIH
2000
Water Quality Studies of Lake Nainital and Surroundings
.
Report no. CS/AR-1/1999-2000, National Institute of Hydrology
,
Roorkee, India
.
Perdikaki
K.
&
Mason
C. F.
1999
Impact of road run-off on receiving streams in eastern England
.
Water Research
33
(
7
),
1627
1633
.
Plum
V.
,
Dahl
C. P.
,
Bentsen
L.
,
Petersen
C. R.
,
Napstjert
L.
&
Thomsen
N. B.
1998
The Actiflo method
.
Water Science and Technology
37
,
269
275
.
US EPA
1993
Handbook: Urban Runoff Pollution Prevention and Control Planning
.
EPA 625-R-93-004, US Environmental Protection Agency
,
Washington, DC, USA
.
Vipat
V.
,
Singh
U. R.
,
Billore
S. K.
2008
Efficacy of rootzone technology for treatment of domestic wastewater: field scale study of a pilot project in Bhopal (MP), India
. In:
Proceedings of Taal 2007, the 12th World Lake Conference
(
Sengupta
M.
&
Dalwani
R.
, eds), pp.
995
1003
.
Weiss
P. T.
,
Gulliver
J. S.
&
Erickson
A. J.
2005
The Cost and Effectiveness of Stormwater Management Practices
.
Report no. MN/RC-2005-23
,
Minnesota Dept of Transportation
,
St Paul, MN, USA
.
WHO
2004
Water Treatment and Pathogen Control: Process Efficiency in Achieving Safe Drinking Water
(
Le Chevallier
M. W.
&
Au
K.-K.
, eds).
World Health Organization and IWA Publishing
,
London
,
UK
.
Yang
K.
,
Li
Z.
,
Zhang
H.
,
Qian
J.
&
Chen
G.
2010
Municipal wastewater phosphorus removal by coagulation
.
Environmental Technology
31
(
6
),
601
609
.
Yu
S.
,
Barnes
S.
&
Gerde
V.
1993
Testing of Best Management Practices for Controlling Highway Runoff
.
Report no. FHWA/VA-93-R16, Virginia Transportation Research Council, Springfield, VA, USA.