PDF 327kB - QUT ePrints

1 downloads 0 Views 359KB Size Report
constructed new or after back-wash, are subjected to a phase of stabilization. Experiments show that bioretention basins are similarly impacted by intermittent ...
Reaching out to the Regions - STORMWATER QUEENSLAND Conference, Townsville, 2014

Stabilization of experimental bioretention basins during intermittent wetting and drying Subramaniam, D.1*, Egodawatta, P.1, Mather, P.1., Rajapakse, J. 1 1

Queensland University of Technology

*Corresponding author email: [email protected]

ABSTRACT Stormwater bioretention basins are subjected to spontaneous intermittent wetting and drying, unlike water treatment filter systems that are subjected to continuous feed. Drinking water filters when constructed new or after back-wash, are subjected to a phase of stabilization. Experiments show that bioretention basins are similarly impacted by intermittent wetting and drying. The common parameter monitored in the stabilisation of filters is the concentration of total solids in the outflow. Filter media in bioretention basins however, consists of a mix of particulate organic matter and fine sand. Organic carbon and solids are therefore needed to be monitored. Four Perspex bioretention filter columns of 94 mm (ID) were packed with a filter layer (800 mm), transition layer and a gravel layer and operated with synthetic stormwater in the laboratory. The filter layer contained 8% organic material by weight. A free board of 350 mm provided detention storage and head to facilitate infiltration. Synthetic stormwater was prepared by adding NH4NO3 (ammonium nitrate) and C2H5NO2 (glycine) and a mixture of kaolinite and montmorillonite clay, to tapwater. The columns were fed with synthetic stormwater with different Antecedent Dry Days (ADD) (0 – 25 day) and constant inflow concentration (2 ppm: nitrate-nitrogen, 1.5 ppm: ammonium-nitrogen, 2.5 ppm: organic-nitrogen 100 ppm: total suspended solids and 7 ppm: organic carbon) at a feed rate of 100mL.min (85.7cm/h). Samples were collected from the outflow at different time intervals between 2 – 150 min from the start of outflow and were tested for Total Suspended Solids (TSS) and Total Organic Carbon (TOC). Both TSS and TOC concentrations in the outflow were observed to be much higher than the concentration of both the parameters in the inflow during the stabilisation period indicating a phase of wash-off (first flush) which lasted for approximately 30 min for both parameters at the beginning of each storm event. The wash-off of TSS and TOC were found to be highly variable depending on the age of the filter and the number of antecedent dry days. The duration of stabilisation phase in the experiments is significant compared with many of the stormwater events. A computational analysis on total mass of each pollutant further affirmed the significance of the first flush of an event on removal of these pollutants. Therefore, the kinetics of the first flush in the stabilisation phase needs to be considered in the performance analysis of the systems.

KEYWORDS Bioretention Basins; Stabilization; Total Suspended Solids; Total Organic Carbon Subramaniam, D., Egodawatta, P., Mather, P., Rajapakse, J., Stabilisation of bioretention basins during intermittent wetting and drying1

Reaching out to the Regions - STORMWATER QUEENSLAND Conference, Townsville, 2014

REQUESTED PRESENTATION FORMAT (delete incorrect option below): 

1

Standard 15 minute Presentation and 5 minute Q&A or

INTRODUCTION

A bioretention basin is a stormwater management device capable of mitigating the impacts of urbanization on quality and quantity of stormwater runoff. They are engineered to remove suspended solids and a range of other water bound pollutants including heavy metals and nutrients. Pollutants in stormwater can create adverse impacts to aquatic environments in receiving waters (Bratieres et al., 2008; Davis et al., 2006; Davis et al., 2003; Hatt et al., 2009; Li and Davis, 2009). In this context, research directed at improving the performance of bioretention basins to reduce pollutant levels are of critical importance for protecting urban environments. Most studies that have examined the performance of bioretention basins to reduce pollutant levels to date have been based on data collection and analysis that have examined event mean concentrations (Bratieres et al., 2008; Davis et al., 2006). Operation of bioretention basins however, occurs sporadically and is irregular during and between rainfall events. Therefore analysis of their performance that depend on the average quality of inflows and outflows, may provide misleading outcomes. This is primarily due to the possibility of stabilization of bioretention basin filters, similar to ripening of sand filters in water treatment systems following a new installation or a backwashing operation. While removal of pollutants in bioretention basins has been widely studied in the past, less attention has been given to the impact of the stabilization of bioretention filters on overall outflow quality even though washoff of pollutants from the bioretention filters itself, most importantly solids and phosphates, has been recorded in such studies (Hsieh et al., 2007). Bioretention basins, unlike sand filter in water treatment systems, operate in two distinctive phases, 1. wet-phase: during the rainfall event and 2. dry-phase: between two rainfall events when there is no inflow. Previous studies have shown that some water is retained in the basin between events. For example, where the concentration of pollutants was higher in the outflow than in the inflow, some authors have observed a decrease in the total mass of pollutants in the outflow owing to the fact that a significant amount of inflow can be retained in the filter itself at the end of that event (Bratieres et al., 2008; Davis et al., 2006; Davis et al., 2003). The impact of any retained water during the dry-phase of an event on pollutant dynamics in the outflow however, has rarely been examined in the past. The first-flush (initial breakthrough and fluctuations of concentration of pollutants in the outflow) that constitute a fraction of the total outflow during the wet-phase of an event could be affected by the water that was retained in the system during the preceding dryphase. The dynamics of fluctuations in concentrations of pollutants in the outflow therefore, could impact the performance of bioretention basins, even when then total mass of pollutants is analysed. During an event the columns operate under stabilized conditions while following a dry-phase, a flush of pollutants from the filter layer itself signifies a destabilized filter. De-stabilisation of filters can occur either due to leaching of material from the filter itself or as a flush of unsettled or disturbed substances from the previous event that had been retained in the filter over the dry-phase. One of Subramaniam, D., Egodawatta, P., Mather, P., Rajapakse, J., Stabilisation of bioretention basins during intermittent wetting and drying2

Reaching out to the Regions - STORMWATER QUEENSLAND Conference, Townsville, 2014

the most common parameters that is monitored for stabilisation of filters of water treatment systems is the concentration of suspended solids in the outflow. Unlike water treatment sand filters, bioretention basin filters consists of engineered sandy loam with added organic matter (Gold Coast City Council, 2003; South East Queensland Healthy Waterways, 2010). While the dynamics of the organic material in bioretention filter has rarely been analysed in the past, because the filter is natural sandy loam, several studies have observed pollutant leaching, most specifically of phosphates (Hsieh et al., 2007). The organic material in the filter layer is therefore, potentially subjected to leaching and thereby can contribute to reduced outflow quality from the system. The current study examined the significance of stabilisation of the filter layer in terms of Total Suspended Solids and Total Organic Carbon on overall pollutant reduction in bioretention basins.

2

METHODOLOGY

2.1

Laboratory-scale bioretention basins

Four Perspex columns each of 94 mm internal diameter and of 1.6 m length were used as experimental bioretention columns (Figure 1 and 2). Each column was packed with granular material according to standard guidelines (Gold Coast City Council, 2003; South East Queensland Healthy Waterways, 2010). Material for bioretention columns (filter media, transition Ponding Zone 300 mm zone and drain zone) was obtained from an industry standard material supplier in Brisbane and the Gold Coast, Australia. Primary characteristics of the bioretention columns are described as below: 1. Filter zone: an engineered filter material of particle size less than 1 mm including 8% of organic material by weight.

Filter Zone

800 mm

2. Transition zone: sand material with particle size ranging from 1 mm to 2 mm, designed to prevent wash off of filter material. 3. Drain zone: gravel material of particle size ranging from 2 mm to 5 mm, serving as the drainage layer.

Transition Zone Drain Zone

4. Ponding zone: an additional ponding zone of 350 mm above the filter layer was left to store stormwater temporarily and also to provide sufficient head for the process of infiltration.

220 mm

Figure 1: schematic diagram of bioretention columns

Columns were fed with tapwater for 2 week (a total of 10 L at each event at a feeding rate of 100 mL/min, two events a week). Column packing was observed to settle and high wash off of filter material was observed during the first few feeds. The packing was designed so that the zones settled to a height as shown in the diagram (Figure 1) within two weeks.

2.2

Experiments

As with the preliminary stabilisation, during experimental runs, columns were fed with synthetic stormwater at a feeding rate of 100 mL/min (equivalent inflow rate per unit area of 850 mm/h) for Subramaniam, D., Egodawatta, P., Mather, P., Rajapakse, J., Stabilisation of bioretention basins during intermittent wetting and drying3

Reaching out to the Regions - STORMWATER QUEENSLAND Conference, Townsville, 2014

approximately 3 hour in each event and were then left to drain. Feed rate was computed based on a 3 month ARI for the Gold Coast region, (45 mm/h rainfall for 30 min) and based on the assumption that the area of the bioretention basin covered an estimated 3% of the total catchment area that is in accordance with the design guidelines (Gold Coast City Council, 2003; South East Queensland Healthy Waterways, 2010). Level of water in the ponding zone above the filter layer was maintained at or below 350 mm. Since columns were not re-packed between events, sequence of events were numbered (EN– event number) to represent the age of the filter represented in fieldscale operations. The experiment was repeated for different number of antecedent dry days (ADD – number of days between two rainfall events) that varied from 0 – 40 day. Columns took approximately 16 Figure 2: The layers of hour to stop draining completely after an event and therefore, zero Bioretention columns antecedent dry day was considered as an event that occurred approximately 24 hour from the previous event. The feed of synthetic stormwater was maintained at a constant strength, comparable to average stormwater quality observed (TN – 5 ppm, TSS – 100 ppm) (Liu, 2011), by adding the following materials to tapwater: • Ammonium nitrate (NH4NO3) and Glycine (C2H5NO2) to represent Total Nitrogen (ammonium-nitrogen, nitrate-nitrogen and organic-nitrogen); •

Kaolinite and montmorillonite clay (1:1 by weight) to represent suspended solids.

Samples were collected from the inflow and tap water during each experimental trial. Additionally, samples were also collected form the outflow stream at 2, 7, 12, 20, 30, 60, 90, 120, 150 min from the beginning of outflow. Samples were monitored for Turbidity (TU) and Total Organic Carbon (TOC) (the focus of this study is on TOC and TSS). A pilot study was conducted to develop a calibration graph between TU and TSS and thus TU observation results were converted to Total Suspended Solids (TSS). Water samples were tested for quality based on Standard Methods for Examination of Water and Wastewater (APHA, 2005).

3

RESULTS

3.1

Total Organic Carbon

Figure 3 show the variation of concentration of Total Organic Carbon (TOC) in the outflow with time. TOC concentration in the outflow fluctuated significantly in the first 30 min of outflow and then declined to reach an asymptote at concentration comparable with inflow concentration. Figure 4 further illustrates the variation in TOC concentrations in the inflow and outflow at different sampling times.

Subramaniam, D., Egodawatta, P., Mather, P., Rajapakse, J., Stabilisation of bioretention basins during intermittent wetting and drying4

Reaching out to the Regions - STORMWATER QUEENSLAND Conference, Townsville, 2014

Figure 3: Concentration of (a) TOC in the outflow with time Figure 4 shows a large variation in the concentration of TOC in the outflow over the first 20 minutes that then gradually declines and reaches asymptote at a constant comparable with inflow TOC concentrations, after approximately 30 min. The inflow had concentration of TOC of approximately 6 – 8 ppm, contributed by both approximately 3 ppm from tapwater and the rest from glycine. This result indicates that there was a first flush of organic carbon in each event irrespective of EN (Event Number), ADD (Antecedent Dry Days) or inflow concentration of TOC (TOCin) and that this lasted for approximately 30 min from the start of outflow. The peak concentrations of TOC in during the first flush however, was observed to be affected by both ADD and EN.

Figure 4: Box and Whiskers plot of TOC concentration in inflow and in outflow at different times

Subramaniam, D., Egodawatta, P., Mather, P., Rajapakse, J., Stabilisation of bioretention basins during intermittent wetting and drying5

Reaching out to the Regions - STORMWATER QUEENSLAND Conference, Townsville, 2014

3.2

Total Suspended Solids

Figure 5 show the variation in concentration of Total Suspended Solids (TSS) in the outflow with time. A similar trend of TOC was also observed for TSS concentration with time, showing a first flush at the beginning of each event in all trials (Figure 5). In contrast to the pattern of TOC concentration with time, there was removal of TSS once the column had stabilized during each event, after approximately 30 min. Another important similarity evident between TOC and TSS was that they both settled, indicating that the filter had stabilized at approximately the same time period from the start of outflow (30 min) for both TOC and TSS.

Figure 6: Cumulative TSS in the inflow and outflow for different events (EN) and antecedent dry days. ADD

In order to understand the impact of first flush on the total mass removal of pollutants in bioretention basins and also to understand the impact of ADD and EN on the overall pollutant removal, Subramaniam, D., Egodawatta, P., Mather, P., Rajapakse, J., Stabilisation of bioretention basins during intermittent wetting and drying6

Reaching out to the Regions - STORMWATER QUEENSLAND Conference, Townsville, 2014

cumulative TSS and TOC were computed based on the inflow and outflow rates and the concentration of TSS and TOC in the inflow and outflow, respectively (Figures 6 and 7). Analysis of the flow rate of the columns (data not shown) showed a lapse of approximately 20 – 30 min was evident between the start of inflow and beginning of outflow that also varied depending on ADD. For this analysis however, it was assumed to be 30 min across all events irrespective of different ADD’s. The time scale for the plots begins with the start of outflow, and therefore, 30 min of inflow was added to the beginning of the cumulative inflow of TSS and TOC. Similarly, the inflow was assumed to have stopped 30 min prior to the end of outflow. It was evident from the cumulative TSS analysis, that cumulative mass removal of TSS in the system was enhanced by age of the filter, and was highly impacted by ADD. Until the 4th event, wash off from the filter was greater than cumulative inflow of TSS, and after which outflow concentration was always less than the inflow concentration of individual pollutants. Meanwhile, the wash off of TOC was always greater than cumulative inflow of TOC. The washoff after zero ADD however, was less than inflow TOC concentrations.

Figure 7: Cumulative TOC in the inflow and outflow with different EN and ADD

4

DISCUSSION

Results of the analysis show, that the filter layer in bioretention systems are subjected to a phase of stabilisation at the beginning of each new rainfall event, that lasts for approximately 30 min in relation to both TOC and TSS. While Antecedent Dry Days (ADD) did have an impact on the peak concentration of both TOC and TSSS during the first flush, ADD did not affect the duration of Subramaniam, D., Egodawatta, P., Mather, P., Rajapakse, J., Stabilisation of bioretention basins during intermittent wetting and drying7

Reaching out to the Regions - STORMWATER QUEENSLAND Conference, Townsville, 2014

stabilization of the bioretention filter. Even though no removal of organic carbon was observed during an event at any point of time, removal of solids did occur once the column had stabilized. For both parameters, TOC and TSS, peak concentrations during the first flush were much higher than their initial concentrations in the inflow, while at the end of the event they settle to concentrations either comparable with (TOC), or less than (TSS) inflow concentrations. Addition of organic material to the filter that makes bioretention filters significantly different from sand filters in water treatment systems. The impact of added organic matter on outflow quality under intermittent wetting and drying conditions that is another unique property of bioretention basins, has not been widely studied. The dynamics of organic carbon under fluctuating moisture content has been studied however, in relation to growth of microbial communities assessed by soil respiration rates (Birch, 1958; Fransluebbers et al., 1994; Kieft et al., 1987; Pesaro et al., 2004; Rudaz et al., 1991; Van Gestel et al., 1993). These studies have focused more on the dynamics of the microbial communities under moisture related stress in the microbial functionality on sudden changes in moisture content on wetting, rather than their dynamics during the drying phase. Observations made in these studies indicate higher microbial growth rate that lasts for approximately up to 10 days in systems that had undergone wetting after a dry period, compared with systems that were continuously in a wet-phase (Bloem et al., 1992; De Nobili et al., 2006; Firer and Schimel, 2002, 2003; Franzluebbers, 2000; Pesaro et al., 2004; Saetre and Stark, 2005). The lapse however, between initiation of wetting and reactivation of microorganisms was not recorded. Furthermore, the above mentioned studies have not been conducted on particulate organic matter. Organic carbon in particulate organic matter is complex in structure and microorganisms cannot immediately consume this material. The dynamics of the transformation of particulate organic matter into labile organic carbon therefore, plays a crucial role in the overall removal/production of organic carbon in bioretention basins. The primary mechanism for removal/retention of solids in a filter is via the process of filtration. In addition, water that is retained in the system potentially holds solids in suspension if it was not filtered during infiltration. In addition, solids in suspension can settle while the fluid is stagnant in pores during the drying phase, and hence remain unattached to filter particles. Solid particles both in suspension and that have settled in the pores, may then be washed off in a subsequent wetting event, ending up in the outflow and also contributed to the peak concentration during first flush. In addition, the progression of wetting front can disturb settled solids that further enhances resuspension of solids that eventually get washed off contributing to the first flush phenomenon. Analysis of cumulative TSS and TOC in outflow in comparison with cumulative supply of TSS and TOC in the inflow in bioretention filters indicate that longer wetting-phases during an event can result in removal of TSS while there is only negligible removal of TOC over the experimental duration. Shorter wetting-phase events on the other hand, may result the opposite effect to a longer wetting-phase event. The current study was based on averages of flow rates and concentration. In addition, it did not consider variation in the water retained due to different number of dry days. Therefore, the relative impact of the length of the wetting-phase potentially may be even higher than what is reported here. Rainfall events are mostly short in duration, and therefore the phase of stabilization is the critical phase where evaluating the analysis of performance of pollutant removal in bioretention basins.

Subramaniam, D., Egodawatta, P., Mather, P., Rajapakse, J., Stabilisation of bioretention basins during intermittent wetting and drying8

Reaching out to the Regions - STORMWATER QUEENSLAND Conference, Townsville, 2014

5

CONCLUSION

A phase of stabilization (due to the first flush of pollutants) was observed across all events that lasted for approximately 30 min, irrespective of Antecedent Dry Days (ADD). Stabilization was observed to occur mainly due to washoff of solids and organic carbon from the filter layer itself. Cumulative analysis on the total mass of pollutants in the inflow and outflow reveal that the first flush of pollutants during stabilization is significant in particular for events that had short wet-phase. Analyses of performance of relative pollutant removal efficiency in bioretention basins therefore, should consider the effect of stabilization of the system.

6

REFERENCES

APHA, 2005, Standard Methods for the Examination of Water and Wastewater: Washington DC, American Public Health Association. Birch, H. F., 1958, The effet of soil drying on humus decomposition and nitrogen availablity: Plant and Soil, v. 10, no. 1, p. 9-31. Bloem, J., de Ruiter, P. C., Koopman, G. J., Lebbink, G., and Brussaard, L., 1992, Microbial numbers and activity in dried and rewetted arable soil under integrated and conventional management: Soil Biology and Biochemistry, v. 24, no. 7, p. 655-665. Bratieres, K., Fletcher, T. D., Deletic, A., Alcazar, L., Coustumer, S. L., and McCarthy, D. T., Removal of nutrients, heavy metals and pathogens by stormwater biofilters, in Proceedings 11th international Conference on Urban Drainage, Edinburgh, UK, 2008. Davis, A. P., Shokouhian, M., Sharma, H., and Minami, C., 2006, Water quality improvement through bioretention media: Nitrogen and phosphorous removal: Water Environment Research, v. 78, no. 3, p. 284-293. Davis, A. P., Shokouhian, M., Sharma, H., Minami, C., and Winogradoff, D., 2003, Water quality improvement through bioretention: lead, copper, and zinc removal: Water Environment Research, v. 75, no. 1, p. 73-82. De Nobili, M., Contin, M., and Brookes, P. C., 2006, Microbial biomass dynamics in recently airdried and rewetted soils compared to others stored air-dry for up to 103 years: Soil Biology and Biochemistry, v. 38, p. 2871-2881. Firer, N., and Schimel, J. P., 2002, Effects of drying-rewetting frequency on soil carbon and nitrogen transformations: Soil Biology and Biochemistry, v. 34, p. 777-787. -, 2003, A proposed mechanism for the pulse in carbon dioxide production commonly observed following the rapid rewetting of a dry soil: Soil Science Socity of America Journal, v. 67, no. 3, p. 798-805. Fransluebbers, K., Weaver, R. W., Juo, A. S. R., and Franzluebbers, A. J., 1994, Carbon and nitrogen mineralization from cowpea plants part decomposing in moist and in repeatedly dried and wetted soil: soil Biology and Biochemistry, v. 26, no. 10, p. 1379-1387. Franzluebbers, A. J., 2000, Flush of carbon dioxide following rewetting of dried soil relates to active organic pools: Soil Science Socity of America Journal, v. 64, no. 2, p. 613. Gold Coast City Council, 2003, Goldcoast planning scheme 2003 Policy 11 - Land development guidelines: Goldcoast. Subramaniam, D., Egodawatta, P., Mather, P., Rajapakse, J., Stabilisation of bioretention basins during intermittent wetting and drying9

Reaching out to the Regions - STORMWATER QUEENSLAND Conference, Townsville, 2014

Hatt, B. E., Fletcher, T. D., and Deletic, A., 2009, Hydrologic and pollutant removal performanceof stormwater biofiltration systems at the field scale: Journal of Hydrology, v. 365, p. 310-321. Hsieh, C. H., Davis, A. P., and Needleman, B. A., 2007, Bioretention column studies of phosphorus removal from urban stormwater runoff: Water Environment Research, v. 79, p. 177-184. Kieft, T. L., Soroker, E., and Firestone, M. K., 1987, Microbial biomass response to a rapid increase in water potential when dry soil is wetted: Soil Biology and Biochemistry, v. 19, no. 2, p. 119-126. Li, H., and Davis, A. P., 2009, Water quality improvement through reductions of pollutant loads using bioretention: Journal of Environmental Engineering, v. 135, no. 8, p. 567-576. Liu, A., 2011, Influence of rainfall and catchment characteristics on urban stormwater quality [Doctor of Philosophy: Queensland University of Technology. Pesaro, M., Nicollier, G., Zeyer, J., and Widmer, F., 2004, Impact of soil drying-rewetting stress on microbial communities and activities and on degradation of two crop protection products: Applied and Environmental Microbiology, v. 70, no. 5, p. 2577-2587. Rudaz, A. O., Davidson, E. A., and Firestone, M. K., 1991, Sources of nitrous oxide production following wetting of dry soil: Federation of European Microbiological Societies, v. 85, p. 117-124. Saetre, P., and Stark, J. M., 2005, Microbial dynamics and carbon and nitrogen cycling following rewetting of soils beneath two semi-arid plant species: Oecologia, v. 142, no. 2, p. 247-260. South East Queensland Healthy Waterways, 2010, Construction and establishment guidelines: Swales, Bioretention systems and Wetlands. Van Gestel, M., Merckx, R., and Vlassak, K., 1993, Microbial biomass responses to soil drying and rewetting: the fate of fast- and slo-growing microorganisms in soils from different climates: Soil Biology and Biochemistry, v. 25, no. 1, p. 109-123.

6.1

Author Biographies (Daniel Subramaniam)

Daniel is in the final year of PhD at Queensland University of Technology, researching on solids and nitrogen removal in bioretention systems. He has completed a Bachelors degree (BSc. Eng) at University of Moratuwa, Sri Lanka (2008) and a Masters degree (MSc. Integrated Environmental Studies) at University of Southampton, UK (2009). Daniel worked as an environmental engineer at a consultancy firm (EML Consultants) in Sri Lanka for 8months, before undertaking a PhD in Australia. The focus of his career is on quantitative and qualitative management of water resources, and designing sustainable systems to overcome constraints in accessibility of fresh water, that is a basic need.

Subramaniam, D., Egodawatta, P., Mather, P., Rajapakse, J., Stabilisation of bioretention basins during intermittent wetting and drying10