Task E, 5 th delivery: Final report (English) on the environmental and technical performance of the treatment unit processes

Transcription

1 Silkeborg Municipality October, 2009 Task E, 5 th delivery: Final report (English) on the environmental and technical performance of the treatment unit processes Contract text to which the deliverable refers: Evaluation of plant uptake (Action E). The pollutant removal due to vegetation is quantified by analyzes of plant tissue. The adsorption of colloidal pollutants to plant surfaces as well as the absorption of dissolved pollutants by the vegetation is addressed. Removal of priority pollutants in terms of organic micropollutants, heavy metals and nutrients are verified. Due to the need of the plants to be in full growth, evaluation of plant uptake of pollutants starts one growth season after completion of the facilities, i.e. autumn Evaluation of sedimentation performance (Action E2, E3, E4). The removal of particulate pollutants by sedimentation is assessed based on mass balances between the pond inlet and different water volumes within the pond. Evaluation of filter performance (Action E5, E6, E7). The pollutant removal efficiency of the filter media is tested to verify the removal of priority pollutants in terms of organic micropollutants, heavy metals and nutrients. The performance of the different sorption media is assessed. Evaluation of technologies for draining stormwater from ponds to filter (Action E8, E9, E0). The technologies of extracting water from the ponds are verified and compared in terms of pollutant removal and in terms of hydraulic capacity. /87

2 Contents General characteristics of the facilities... 7 Planting... 8 Odense... 8 Aarhus... Silkeborg... 2 Nutrients and heavy metals in plants... 6 Phosphorus (P)... 6 Potassium (K)... 6 Calcium (Ca)... 6 Sodium (Na)... 6 Iron (Fe )... 6 Manganese (Mn)... 6 Aluminum (Al)... 7 Lead (Pb)... 7 Zink (Zn)... 7 Cadmium (Cd)... 7 Nickel (Ni)... 7 Chromium (Cr)... 7 Copper (Cu)... 7 Discussion of nutrients and heavy metals in plants... 8 Pollutant loadings from the 3 catchments... 9 Overall removal of pollutants... 2 Removal of pollutants at the facility in Odense... 2 Copper... 2 Zinc Lead Cadmium Chromium Nickel Mercury Total Nitrogen Total Phosphorous Orthophosphate /87

3 Total Suspended Solids (TSS) Oil and grease PAH after USEPA Discussion of the effect of fixed media sorption filters Removal of pollutants at the facility in Århus Copper Zinc Lead Cadmium Chromium... 3 Nickel Mercury Total Nitrogen Total Phosphorous Orthophosphate Total suspended solids (TSS) Oil and grease PAH after USEPA Discussion of the effect of iron enrichment of bottom sediments Removal of pollutants at the facility in Silkeborg Copper Zinc Lead Cadmium Chromium Nickel... 4 Mercury... 4 Total Nitrogen Total Phosphorous Orthophosphate Total suspended solids (TSS) Oil and grease PAH after USEPA Discussion of the effect of aluminum addition /87

4 Sediment Nutrients and heavy metals in sediment Odense Phosphorus (P) Potassium (K) Calcium (Ca) Sodium (Na) Iron (Fe ) Manganese (Mn) Aluminum (Al) Lead (Pb) Zinc (Zn) Cadmium (Cd) Nickel (Ni) Chromium (Cr) Copper (Cu) Århus Phosphorus (P) Potassium (K) Calcium (Ca) Sodium (Na) Iron (Fe )... 5 Manganese (Mn)... 5 Aluminum (Al)... 5 Lead (Pb)... 5 Zink (Zn)... 5 Cadmium (Cd)... 5 Nickel (Ni)... 5 Chromium (Cr)... 5 Copper (Cu)... 5 Silkeborg Phosphorus (P) Potassium (K) Calcium (Ca) /87

5 Sodium (Na) Iron (Fe ) Manganese (Mn) Aluminum (Al) Lead (Pb) Zink (Zn) Cadmium (Cd) Nickel (Ni) Chromium (Cr) Copper (Cu) Organic content PAHs in sediment Algae growth in the ponds The removal effectiveness of the wet ponds The removal effectiveness of the sand filters On line measurement of water quality data The facility in Odense The facility in Århus The facility in Silkeborg Hydraulic characteristics of the ponds The facility in Odense Estimation of catchment area using flow measurement The facility in Århus Estimation of catchment area using flow measurement The facility in Silkeborg Estimation of catchment area using flow measurement Mixing and flow pattern in the ponds The effect of stormwater pond compartmentalization... 8 Stormwater toxicity versus pollutant content... 8 The facility in Odense... 8 The facility in Århus Evaluations and conclusions The effectiveness of the advanced treatment technologies Fixed-media sorption /87

6 Iron enrichment of bottom sediments Aluminum addition to incoming stormwater Sand filters Plants Overall conclusions /87

7 General characteristics of the facilities The following report addresses monitoring at the 3 facilities from start of operation until the project closure ultimo September, Three full-scale facilities were constructed to test and demonstrate the technologies for advanced treatment of stormwater runoff. One facility was constructed for each technology (Figure ). All designs were based on wet detention ponds as the first treatment step, and all designs contained sand filters to treat the outlet water. Three different sand filter layouts were tested: One horizontal sand filter, one sloping filter and one vertical filter. Characteristics of the 3 facilities are presented in Table. Figure. The principle layout of the 3 facilities for advanced stormwater treatment Table. Characteristics of the 3 facilities for advanced stormwater treatment Advanced treatment technology Fixed-media sorption Iron enrichment Aluminum addition Location Odense Århus Silkeborg Type of catchment Light industry Residential (blocks of flats) Residential (detached houses) and highway Annual precipitation 657 mm y - 66 mm y - 79 mm y - Total catchment area 27.4 ha 57.4 ha 2.5 ha Impervious catchment area.4 ha 25.8 ha 8.8 ha Permanent wet volume,990 m 3 6,900 m 3 2,680 m 3 Retention volume,300 m 3,400 m 3 3,230 m 3 Permanent water depth, max Area of horizontal sand filter 00 m m 2 80 m 2 Length of sloping sand filter 30 m 65 m 30 m Length of vertical sand filter m 2.6 m 6.3 m ) The slope of the sand filter was :5 and stretched to the maximum water level of the retention volume 2) The height sand filter was to the maximum water level of the retention volume 7/87

8 Planting The plant selection and the planting plan for the three systems was decided according to the physical features of each wet pond, operational characteristics and the esthetical needs to ensure functionality while maintaining a natural and pleasant appearance and to enhance the ecological and educational value of the sites by attracting fauna and to encourage visits by the local public. Each facility had a specific design as follows: the facility in Odense was planted with a mix of larger and smaller littoral helophytes on the banks and with Phragmites australis (common reed) on the sand filters. The banks of the facility in Århus were left unplanted but the sandfilter was planted with Phragmites australis. Three high grounds (polders) planted with Schoenoplectus lacustris (Common Club-rush) were established in the pond to make the pond resemble the neighboring Lake Brabrand. The banks of the facility in Silkeborg were planted with a mix of larger and smaller littoral helophytes and the sand filters and the barriers with Phragmites australis. Samples of all the plant species planted in the system were collected and analyzed for nutrients and heavy metals. The nutrient elements analyzed included phosphorus, potassium, calcium, iron, sodium, manganese, and aluminum. The heavy metals analyzed were lead, zinc, cadmium, nickel, chromium and copper. The plants were collected after at least one year of operation. Depending on the species, their parts were separated into roots, rhizomes, stems and leaves and analyzed individually. For some of the heavy metals, the concentrations in the plant tissue were below the detection limits (see Table 2). When a measurement was below the detection limit, half of the detection limit was assumed in the calculations. Table 2. Detection limits for element analysis of plant materials by ICP-OES (per dry weight basis). Parameter Unit Detection limit Phosphorus (P) mg/g d.w. 0.0 Potassium (K) mg/g d.w Calcium (Ca) mg/g d.w Sodium (Na) mg/g d.w Iron (Fe) mg/g d.w. 0.0 Manganese (Mn) mg/g d.w Aluminum (Al) mg/g d.w Lead (Pb) µg/g d.w. 2 Zink (Zn) µg/g d.w. 2 Cadmium (Cd) µg/g d.w. 0.5 Nickel (Ni) µg/g d.w. 0.5 Chromium (Cr) µg/g d.w. 0.5 Copper (Cu) µg/g d.w. 5 Odense The planting plan for the system at Odense comprises a mixture of plant species that should benefit the performance, as well as the perception of the system among the local residents. The species planted included plants that should help counteract clogging of the sandfilter (Phragmites australis). Other species were selected mainly for aesthetic reasons. Tall and robust plants (Typha sp. and Rumex sp.) were planted along the side facing an industrial area and around outlet structures. Smaller species that can stabilize banks and are also ornamental giving the system a pleasant appearance (e.g. Sagitaria sagitafolia, Alisma lanceolata, Iris pseudacorus, Caltha palustris, Nymphaea/Nuphar sp.) and were planted next to the public walk path passing the system. All the plants were sown in pots and transplanted to the site once the system 8/87

9 was commissioned and after the plants had grown a growing season in the pots. Figure 2 shows the planting scheme for the system in Odense. Figure 2. Planting plan for the system in Odense Table 3 presents average concentrations of elements in Phragmites australis collected from the facility at Odense and Table 4 the concentration in other species Table 3. Average concentrations of nutrients and heavy metals in Phragmites australis tissues from the facility in Odense July Parameter Unit N Roots Rhizomes. Stems Leaves Ave*. Std** Ave*. Std** Ave*. Std** Ave*. Std** Phosphorus (P) mg/g d.w Potassium (K) mg/g d.w Calcium (Ca) mg/g d.w Sodium (Na) mg/g d.w Iron (Fe) mg/g d.w Manganese (Mn) mg/g d.w Aluminum (Al) mg/g d.w < < Lead (Pb) µg/g d.w <2 - <2 - Zink (Zn) µg/g d.w Cadmium (Cd) µg/g d.w 3 <0.5 - <0.5 - <0.5 - <0.5 - Nickel (Ni) µg/g d.w Chromium (Cr) µg/g d.w Copper (Cu) µg/g d.w * Average ** Standard deviation 9/87

10 Table 4. Average concentrations (±. SD) of nutrients and heavy metals in plant tissues from the facility in Odense July 2009 (BDL = Below Detection Limit). Plant Species Plant Tissue P Fe Mn Ca Na K Al Pb Zn Cd Ni Cr Cu mg g - d.w. µg g - d.w. Alisma lanceolatum whole plant BDL Iris pseudacorus bulks 2.5±0.7.9± ±0.05 2± ±0.7 2± ± ±2. 73±54 BDL 2.±.8 5.9±5.0 62±64 leaves 4.3±0.2 0.± ±0.2 22±9.4.±0.2 26±8 BDL BDL 8±4.2 BDL BDL.3±0.2 4±4. Ranunculus lingua whole plant BDL Rumes hydrolapathum roots BDL ,700 BDL ,400 storage roots BDL 00 BDL leaves BDL 92 BDL Sagittaria sagittifolia whole plant BDL Typha angustifolia roots 3.3±.2 23±2.2.0±0.03 5±6.6 2±0. 3± ±0.5 46±9.3,30±300 BDL 49±9.5 2,90±390 rhizomes 6.5± ± ± ± ±.2 28± ±0.05.5±0.7 0±28 BDL 4.9±.0.6±0.3 8±39 leaves 5.3±0.3 0.± ± ±0.5 4.±0.03 9±.3 BDL BDL 30±5.0 BDL 2.±0.5.3±0.9 ±0.27 Typha latifolia roots 4.5±0.7 22±.0 0.7±0.2 4±.9 3±0.7 20±.4.7±0. 49±0,080±57 BDL 46±5.0 2±.5,70 rhizomes 8.± ±2.2 0.± ± ±2.9 32± ± ±2.0 55±30 BDL 5.8± ±0.6 00±43 leaves 5.4± ± ± ± ±0.4 27± ±0.0 BDL 4±8.6 BDL 3.4±0.3.0±0.4 9±2.4 Typha minima roots BDL rhizomes BDL leaves BDL /87

11 Aarhus The stormwater treatment system in Aarhus is located in the western part of the city, adjacent to a relatively large natural lake (Brabrand lake), whereto the treated run-off waters are discharged. The planting scheme was designed to minimize environmental and visual impacts and with regard to the performance of the sandfilters. Species compatible with the already existing plants in the recipient lake Brabrand were selected in order to reduce the risk of spreading of invasive species and to make the system visually harmonious with the lake. The selected plants were Phragmites australis (common reed), Schoenoplectus lacustris (club-rush) and Nymphaea alba (water lily). Phragmites australis was planted in the sandfilter section to counteract erosion and maintain the hydraulic conductivity of the sandfilter by its deep growing root system. Schoenoplectus lacustris were planted in three polders built in the wet pond to mimic the Schoenoplectus lacustris polders present in Lake Brabrand. The polders should minimize the impact for bird species and the possible visual impact created by the new structure. The Nymphaea alba was planted in the deepest part of the pond but did not survive. Table 5 presents the results of elements measured in Phragmites australis and Error! Reference source not found. the concentration in Schoenoplectus lacustris. Table 5. Average concentration of nutrients and heavy metals in Phragmites australis tissues collected from the facility in Aarhus May, Parameter Unit N Roots Rhizomes. Stems Leaves Ave*. Std** Ave*. Std** Ave*. Std** Ave*. Std** Phosphorus (P) mg/g d.w Potassium (K) mg/g d.w Calcium (Ca) mg/g d.w Sodium (Na) mg/g d.w Iron (Fe) mg/g d.w Manganese (Mn) mg/g d.w Aluminum (Al) mg/g d.w Lead (Pb) µg/g d.w <2 - <2 - <2 - Zink (Zn) µg/g d.w Cadmium (Cd) µg/g d.w 3 <0.5 - <0.5 - <0.5 - <0.5 - Nickel (Ni) µg/g d.w <0.5 - <0.5 - Chromium (Cr) µg/g d.w Copper (Cu) µg/g d.w <5 - <5 - <5 - * Average ** Standard deviation /87

12 Table 6. Average concentration of nutrients and heavy metals in Schoenoplectus lacustris tissues collected from the facility in Aarhus May, Parameter Unit N Roots Rhizomes. Leaves Ave*. Std** Ave*. Std** Ave*. Std** Phosphorus (P) mg/g d.w Potassium (K) mg/g d.w Calcium (Ca) mg/g d.w Sodium (Na) mg/g d.w Iron (Fe) mg/g d.w Manganese (Mn) mg/g d.w Aluminum (Al) mg/g d.w Lead (Pb) µg/g d.w. 3 <2 - <2 - <2 - Zink (Zn) µg/g d.w Cadmium (Cd) µg/g d.w 3 <0.5 - <0.5 - <0.5 - Nickel (Ni) µg/g d.w Chromium (Cr) µg/g d.w Copper (Cu) µg/g d.w 3 5 <5 - <5 - Silkeborg The system in Silkeborg was established where no system was present before and gave the possibility of an entire new concept, including the geometrical and physical design with two transversal sand filtering sections where the stormwater at low water flow is filtered through. At high water flow, when the water level rises, the filter sections are flooded. The system is located in a natural area, and planting were done with consideration to the system operation and so that the system will appear as a 'natural' habitat within the landscape. The transversal sandfilters and outlet filters were planted with Phragmites australis because of its deep and dense root system that counteract erosion and helps to maintain a high hydraulic permeability of the filter. Twelve other plants species were planted on the banks as well as in the flooded sections. On the banks between the filter sections large and robust marsh plants and marsh plants with lower growth and beautiful flowers are planted. Water lilies were planted in the last section before the outlet. The free-floating plant Stratiotes aloides was also planned for the last section, but these did not develop. Plant growth has in general, except the Phragmites australis, been delayed because of a delayed commissioning, but after one growing season all the plants have survived. The plants were sampled for nutrient and heavy metal analysis after one year of operation. Table 7 presents average concentration of elements measured in Phragmites australis, and table 8 the concentration in other species 2/87

13 Figure 3. Planting plan for the system in Silkeborg 3/87

14 Table 7. Average concentration of nutrients and heavy metals concentrations in Phragmites australis from tissues sampled at the facility in Silkeborg August, Parameter Unit N Roots Rhizomes. Stems Leaves Ave*. Std** Ave*. Std** Ave*. Std** Ave*. Std** Phosphorus (P) mg/g d.w Potassium (K) mg/g d.w Calcium (Ca) mg/g d.w Sodium (Na) mg/g d.w Iron (Fe) mg/g d.w Manganese (Mn) mg/g d.w Aluminum (Al) mg/g d.w Lead (Pb) µg/g d.w <2 - <2 - <2 - Zink (Zn) µg/g d.w Cadmium (Cd) µg/g d.w 9 <0.5 - <0.5 - <0.5 - <0.5 - Nickel (Ni) µg/g d.w Chromium (Cr) µg/g d.w Copper (Cu) µg/g d.w <5 - <5 - <5 - * Average ** Standard deviation 4/87

15 Table 8. Average (±SD) concentration of nutrients and heavy metals concentrations in plants from tissues sampled at the facility in Silkeborg August, 2009 (BDL = Below Detection Limit). Plant Plant Tissue P Fe Mn Ca Na K Al Pb Zn Cd Ni Cr Cu mg g - d.w. µg g - d.w. Alisma lanceolatum whole plant 3± 7.6± ±.0 7.7±0.5 0±2.3 63±3 2.3±.2 BDL 4±42 BDL 5.±.6 7.7±3.0 9±6.4 Caltha palustris roots 4.7±.2 23± ±.8 2± ± ± ± ± ±89 BDL 6.2±.5 ±.9 25±6.0 leaves 2±2.5.6±0.5 2.±0.9 0±.4 ±6.6 52±5.3.±0.2 BDL 32±63 BDL 2.4± ±.6 ±4.9 Iris pseudacorus roots 4.±.2 4.3± ± ±.0 3.7±.2 3±4.3.6±0.7 BDL 75±5 BDL 2.8±.0 5.4±2. ±3.9 bulks 4.0±.7 0.9± ±0. 0± ±0.8 27±2 0.6±0.5 BDL 43±3 BDL.9± ±4. 4±5.7 leaves 5.7± ± ±0.3 2±2.8.6±0.2 34± ±0.3 BDL 28±6.6 BDL.3±0.6 2.± ±2.2 Ranunculus lingua roots 3.5±0.8 9±7.9±.3 0± ± ± ±2. 2.0±.6 53±23 BDL ±2.8 38±7 27±8.6 rhizomes 3.8± ± ±0.2 7.±.8 2.7±. 7.0±2.6.9±.2.3±0.5 30±2 BDL 2.2±.6 4.8±2. 8.8±.9 leaves 4.7±0.6.7±0.7 3.±0.4 2±3.7 9.±.8 5±3.2.4±0.7 BDL 95±3 BDL 2.5±.2 4.6± ±3.7 Rumex hydrolapathum roots 3.3±0.6 ±9 2.±.8 9,8±.6 5.6±2.7 3± ±.4 BDL 40±66 BDL 5.± ±3.2 25±7.5 storage roots 3.± ± ±.0.5±.8 8.3±.5 0.4±0.8 BDL 32±22 BDL.±.5.7±3.3 5±4.0 leaves 3.0± ±0. 0.7±0.2 8± ±0.9 7±0.7 0.±0. BDL 8±4.3 BDL 0.3±0.2.0± ±3.4 Saggittaria sagittifolia whole plant 8.6±.5 0± ±.3 8.±.3 9.9±3.9 37±0 3.0±0.6.9±.5 7±34 BDL 6.5±0.6 6±6.5 5±6.9 Scirpus maritimus roots 3.8±.6 32±2.4± ±0. 6.4± ± ±.2 2.2±.2 6±3.4 BDL 6.7±0.9 24± 7±0,5 rhizomes 4.8±.5 3.4± ± ± ±0.9 7± ±0.3 BDL 29±5.3 BDL 4.±3.7 3± 7.7±2.0 leaves 4.7±.4 0.7± ± ± ±2.7 8± ±0.3 BDL 25±8.3 BDL 0.9± ±. 6.0±3. Sparganium erectum roots 4.0±0.9 29±.2 6.3± ±.5 5.2±0.9 6± ±0.2 BDL 38±24 BDL 7.7±3.9 6±7. ±3.8 rhizomes 4.2±0.3 2±3.3±.4 4.9±. 6.9±.2 28±.7 0.6±0.5 BDL 35±0 BDL 0.6± ±3.0 5±2.2 leaves 8.±.0 2± ±.5 ±2.9 ±.6 33±7.0.2±0.5 BDL 6±9 BDL 2.6±.7 3,9±2.3 2±2.0 Stratiotes aloides whole plant BDL 35 BDL ±0.0 Typha angustifolia roots 4.2±0.7 5±3.9.9± ±0.4 9.±3.5 25± ±3.4.7±.6 56±66 BDL 6.4±4.5 ±6.5 9±4 rhizomes 6.4±2. 2.4±.9 0.3±0. 4.7±0. 5.8±2.0 36±5 0.4±0.4 BDL 22±8.8 BDL.±0.8 3.± ±2.2 leaves 5.5±.0 0.3± ± ± ±.4 38±7.9 0.±0. BDL 6±4.4 BDL 0.9±0.3.2± ±2.3 Typha minima roots 3.7±.3 5±6.3.0±0.6 0± ± ± ±4. BDL 65±62 BDL 4.8± ±6.2 ±4.9 rhizomes 4.6±.7 3.±.6 0.4±0.2 3± ±4.2 7±22 0.7±0.2.4±0.7 57±4 BDL.8±0.9 2± ±6.6 leaves 5.0± ±0. 0.8± ±.2 6.9±2.0 30±5.2 0.±0. BDL 24±4.7 BDL 0.5± ±.3 5.9±2.9 5/87

16 Nutrients and heavy metals in plants Phosphorus (P) The concentration of P varies among sites and plant tissues with the highest concentrations of P in the leaves. The P concentrations in the Phragmites australis sampled from the three systems show that the concentrations in plants from Silkeborg and Odense are within expected P concentrations reported in the literature for plants growing in natural environment (between 0.4% to 0.30% dry weight), while the P concentration measured in plants sampled from Århus is higher than in natural systems but corresponding to plants sampled in agricultural drainage areas. Among other plants sampled from the systems, Alisma lanceolatum (narrow-leaved water plantain) showed the highest concentration of P with similar concentrations in plants sampled at Silkeborg and Odense. Potassium (K) Potassium concentrations in Phragmites tissues sampled from the sites are higher in the aerial tissues than other plant parts. Stems sampled at Århus had the highest concentration of K (29 mg/g d.w.) which is about twice the concentrations in stems from the other two systems. Alisma lanceolatum had K concentrations of around 50 mg/g d.w. in samples collected from both systems. Calcium (Ca) Calcium concentrations are highest in the roots of Phragmites australis in all systems with similar concentrations in Århus and Odense (0 mg/g d.w.) and slightly lower concentrations in plants sampled at Silkeborg. In other plant species, the Ca concentrations did not vary much between plants parts and systems. Sodium (Na) Sodium concentrations in Phragmites australis are similar for all part tissues and sites with the highest concentrations in the roots, ranging from 4.7 to 6.3 mg/g d.w. Alisma lanceolatum (narrow-leaved water plantain) in the system in Silkeborg had average concentrations of around 2 mg/g d.w. Iron (Fe ) Iron concentrations are, as expected, generally highest in roots of the plants. For Phragmites australis the concentrations in the roots ranged from ca. 0 mg/g d.w in Silkeborg to 22 mg/g d.w. in Århus (iron salts were added to the system). The iron concentrations in the aerial parts of the plants are similar for the three systems, around.0 mg/g d.w. for stems and no higher than.0 mg/g d.w. in the leaves. The Fe concentrations in roots of Phragmites australis in the three systems are slightly higher than reported in the literature for reeds growing in natural environments, but concentrations in stems and leaves are at the same level. The other plant species planted in the systems showed a similar pattern with higher concentrations in the roots. Manganese (Mn) Manganese concentrations are highest in the Phragmites australis root in all the systems ranging from 0.7 in Silkeborg to 7.2 mg/g d.w. in Århus. The manganese concentrations of the aerial parts of Phragmites australis from the three systems never exceeded.0 mg/g d.w. 6/87

17 Aluminum (Al) The highest average concentrations of Al were found in the roots of the plants. In Phragmites australis in Århus, Al concentration were ca µg/g d.w., and in Silkeborg and Odense ca µg/g d.w. In other plant parts, Al concentrations were well below 000 µg/g d.w. The Al concentrations in reed roots in Silkeborg and Odense are comparable to those in reeds in natural habitats, while the Al concentrations in Århus are about twice as high. The Al concentrations in the aerial tissues of Phragmites australis are similar to those in reeds from natural vegetations. Other plant species also had the highest Al concentration in the roots, and in some species the Al concentration in leaves was below the detection limit (in Odense for Typha leaves). Lead (Pb) Concentrations of lead in the plant tissues were generally low and close to the detection limit of the analytical procedure (2 µg/g d.w.). However, in root of P. australis in the Odense system the Pb concentrations were particularly high (82 µg/g d.w.) and approximately 20 times higher than in Århus and Odense. In other species similar concentrations of Pb were found in the roots, but in aerial tissues concentrations were low. The Pb concentrations in leaves of P. australis from the three systems are similar to those in leaves from natural vegetations. Zink (Zn) Zn is a heavy metal found frequently in urban runoff, and particularly in the Odense system, the concentrations of Zn in the water were high (see pollutant loading). The Zn concentration in roots of Phragmites australis differed greatly between systems, probably due to the loading (Odense: 900 µg/g d.w., Århus 300 µg/g d.w. and Silkeborg: 00 µg/g d.w.). The concentrations of Zn in the aerial tissues of Phragmites were much lower, but still the concentrations at Odense were about twice as high (ca. 0 µg/g d.w.) as concentrations at Århus and Silkeborg (40-50 µg/g d.w.). In other plant species, Zn concentrations were also consistently higher in the roots. Rumex roots collected at Odense had Zn concentrations as high as 4700 µg/g d.w., but low concentrations in the leaves (90 µg/g d.w.) Cadmium (Cd) Cadmium was always under detection limit (0.5 µg/g d.w.) for all the plants sampled from the three sites. Nickel (Ni) Nickel concentrations in roots of Phragmites tissues are significantly higher in Odense (00 µg/g d.w.), than in Århus (40 µg/g d.w.) and Silkeborg (0 µg/g d.w.), and higher than in Phragmites roots in natural vegetations. The Ni concentrations in Phragmites leaves and stems are around 0 times lower than the Ni concentrations in the roots. Ni concentrations in the other plant species are relatively low. The highest concentration measured was in Rumex roots from Odense (ca. 200 µg/g d.w.) Chromium (Cr) The concentration of Cr in roots of Phragmites tissues did not differ much between the systems (Odense: 3 µg/g d.w., Århus 9 µg/g d.w. and Silkeborg: 0 µg/g d.w.) The concentrations of Cr in the aerial parts are at least 0 times less than the concentrations in roots. In other plant species, the Cr concentration is similar to that of Phragmites, except for Schoenoplectus roots in Århus with a Cr concentration of 50 µg/g d.w. Copper (Cu) Copper concentrations in the water at Odense was high (see water quality section), and therefore the plants in the Odense system have been exposed to high Cu concentrations. The Cu concentrations 7/87

18 measured in the Phragmites tissues collected from the systems show that the Cu is found mainly in the roots with Cu concentrations of 2000, 20 and 0 µg/g d.w. for Odense, Århus and Silkeborg, respectively. Hence, Cu concentrations in roots are 200 times elevated in Odense compared to Silkeborg. Cu in the other tissues of Phragmites is at least a factor 0 lower than in the roots (for Phragmites from Odense more than a factor 35). In other plant species, the same trend is observed. Plants from Odense have higher Cu concentrations. Particularly Rumex roots had high Cu concentrations (ca µg/g d.w.). The Cu concentrations in plants from Århus and Silkeborg are comparable to the concentrations reported for natural vegetations. But the Cu concentrations in the roots in Odense are at least a factor 00 higher than concentrations in roots sampled from natural plant stands. Discussion of nutrients and heavy metals in plants The elements analyzed are all macro- or micronutrients for plants (except Al, Pb and Cd), and as such these elements are needed for the proper growth of the plants. Macronutrients (N, P, K, Ca, Mg, S, Si) occurs in concentrations in the plant tissues that are more than 0 times higher than micronutrients (Fe, Mn, Mo, Zn, Cu, Ni, B, Cl, Na). Hence, plants have mechanisms to take up and process these ions in their tissues. Also trace elements, that are not needed by the plants, such as Pb and Cd, are taken up, but plants may have strategies or mechanisms to minimize the uptake by roots and translocation within the plant. Macronutrients, such as N, P and K, are taken up by the roots and efficiently transported to the leaves where they are needed in high quantity. The concentration of these elements is therefore naturally higher in leaves than in roots and rhizomes. Micronutrients, such as Fe, Mn, Cu and Zn, are needed in much lower quantities. Hence, transport from roots to leaves is much less, and concentrations in leaves might be lower than in roots and rhizomes. Many trace metals, such as Pb, are barely transported from the roots to the leaves, and hence once taken up by the roots, they remain in the roots. The nutrient and heavy metal concentrations in the plant tissues sampled in the three systems are thus dependent on the physiological requirements of the plants (concerning macro and micronutrients) and the external concentrations that the plants are exposed too. In general, the element concentrations found in the Phragmites stems and leaves from the three systems are comparable to concentrations in plants growing in natural unaffected areas. Heavy metals are taken by the roots of the plants, and particularly Cu and Zn were accumulated in the roots of the plants in the Odense systems. However, the heavy metals were barely translocated from the roots, and the concentrations in the stems and leaves were much less elevated. This is of importance when considering the risk of spreading and potential biomagnifications of heavy metals in the food chain. The plant tissues (leaves, stems) that are eaten by birds and insects generally contain much lower concentrations of the hazardous heavy metals compared to the roots. Hence, the risk of spreading and biomagnifications of heavy metals from the ponds through plant uptake and subsequent grazing is very limited. The hazardous heavy metals taken up by the plants remains in the roots, and upon dead and decomposition of the roots, they will be bound in the sediment. The concentrations of heavy metals are consistently higher in plants sampled from the system in Odense than Århus and Silkeborg, as a consequence of the high concentrations in the water and the sediment. Concentrations of some heavy metals (Zn, Cu, Ni ) in roots are a factor 00 times higher than in roots of plants growing in natural unaffected vegetations. Although the concentrations of heavy metal in plants in the Odense system are significantly elevated in roots, the concentrations in the aerial tissues from the same plants are low and do not generate threats for the fauna or visitors to the site. 8/87

19 Pollutant loadings from the 3 catchments The stormwater runoff from the catchment in Odense was the most polluted of the 3 catchments, followed by the catchment in Århus. The catchment in Silkeborg contributed the comparatively least polluted runoff (Table 4, Table 5, Table 6). The average values in the tables are flow weighted average concentrations. When a measurement was below the detection limit, half of the detection limit was assumed in the calculations. Table 4. Pollutant concentrations in the inlet to the facility in Odense Parameter Unit N N >det. Average Median 25% 75% limit flowprop. percentile percentile Sum PAH (6EPA) µg/l Lead (Pb) µg/l Cadmium (Cd) µg/l Chromium (Cr) µg/l Copper (Cu) µg/l Mercury ( Hg) µg/l Nickel (Ni) µg/l Zink (Zn) µg/l Total Suspended Solids (TSS) mg/l Total N mg/l Orthophosphate-P mg/l Total P mg/l Oil and fat mg/l Volatile suspended solids (VSS) % % 37% 35% 39% Chloride mg/l Alkalinity mmol/l Chemical oxygen demand (COD) mg/l Iron mg/l Aluminum µg/l /87

20 Table 5. Pollutant concentrations in the inlet to the facility in Århus Parameter Unit N N >det. limit Average flowprop. Median 25% percentile 75% percentile Sum PAH (6EPA) µg/l Lead (Pb) µg/l Cadmium (Cd) µg/l Chromium (Cr) µg/l Copper (Cu) µg/l Mercury ( Hg) µg/l Nickel (Ni) µg/l Zink (Zn) µg/l Total Suspended Solids (TSS) mg/l Total N mg/l Orthophosphate-P mg/l Total P mg/l Oil and fat mg/l Volatile suspended solids (VSS) % 5 5 5% 42% 39% 73% Chloride mg/l Alkalinity mmol/l Chemical oxygen demand (COD) mg/l Iron mg/l Aluminum µg/l Table 6 Pollutant concentrations in the inlet to the facility in Silkeborg Parameter Unit N N >det. limit Average flowprop. Median 25% percentile 75% percentile Sum PAH (6EPA) µg/l Lead (Pb) µg/l Cadmium (Cd) µg/l Chromium (Cr) µg/l Copper (Cu) µg/l Mercury ( Hg) µg/l Nickel (Ni) µg/l Zink (Zn) µg/l Total Suspended Solids (TSS) mg/l Total N mg/l Orthophosphate-P mg/l Total P mg/l Oil and fat mg/l Volatile suspended solids (VSS) % Chloride mg/l Alkalinity mmol/l Chemical oxygen demand (COD) mg/l Iron mg/l Aluminum µg/l /87

21 The runoff from the catchment in Odense was characterized by rather large amounts of heavy metals, especially copper was high, however zinc, nickel and lead were also high compared to the two other facilities. The highest concentration of copper, zinc and lead in the runoff was 3,300 mg m -3, 2,00 mg m -3 and 0 mg m -3, respectively. The maximum concentration of copper was more than 00 times higher than the maximum concentrations of the runoff water from the catchments in Silkeborg and Århus. Zinc and lead were about 0 times higher. The increased concentrations of these metals correlated, and it is therefore highly likely that they originate from the same source, which probably is an illicit industrial discharge of these heavy metals. The catchment in Århus was characterized by a certain amount of false connection were sanitary wastewater was discharged to the stormwater system. At times the bacterial load (E-Coli and Coliforms) was high and at other times no bacteria were present. At the inlet, wastewater odor could at times be found and grab samples from the inlet pipe showed up till 35,000 Total Coliform bacteria per 00 ml and E. Coli per 00 ml. At other times the bacteria counts were zero. The origin of these sporadic wastewater discharges is unknown. The catchment in Silkeborg seemed not to receive any undesired wastewaters and pollutants. The pollutant monitoring cover the periods from March 2008 to September 2009 (Odense), from June 2008 to September 2009 (Århus), from December 2008 to September 2009 (Silkeborg). Overall removal of pollutants The removal of pollutants is illustrated for each facility. Here after the different technologies are compared with respect to treatment efficiency. Removal of pollutants at the facility in Odense The facility for testing fixed-media sorption (Odense) was taken into operation in February It consisted of a main filter filled with 55 m 3 of Oyta Shells (Oytaco Ltd, Denmark), a natural product obtained from large deposits of fossil oyster shells in the shallow waters of the North Sea. The composition of the material was 96% CaCO 3 and MgCO 3 with a calcium content of 38%. The water drained through the sand filters and through the sorption filter by gravitation. In addition to the main filter, the facility was equipped with 3 test filters fed by intermittent pumping. One held 2.5 m 3 of Oyta Shells, another held 2.5 m 3 of granulated olivine (Filtersil 2749 from North Cape Minerals, Norway). The last filter was constructed as a sandwich of 0.5 m 3 of Oyta Shells as the bottom layer on top of which was 0.5 m 3 of iron oxide coated olivine (Filtersil TOC from North Cape Minerals, Norway) and.5 m 3 of Oyta Shells. Copper The concentration of copper in the inlet to the facility in Odense was rather high compared to typical Danish stormwater (typical Danish stormwater has a copper median content around 20 μg/l). The copper was removed down to an average concentration of 4 μg/l, corresponding to an overall removal rate of 99% (Figure 4). 2/87

22 Copper [mg m -3 ] , After sorption Figure 4. Removal of copper at the facility in Odense. The grey line illustrates the average values of the measurements Zinc The concentration of zinc in the inlet to the facility in Odense corresponded to typical Danish stormwater, even though it was in the higher range hereof (typical Danish stormwater has a zinc median content around μg/l). The zinc was removed down to an average concentration less than the detection limit of 5 μg/l, corresponding to an overall removal rate better than 99% (Figure 5) Zinc [mg m -3 ] 00 0 After sorption Figure 5. Removal of zinc at the facility in Odense. The grey line illustrates the average values of the measurements Lead The concentration of lead in the inlet to the facility in Odense corresponded to typical Danish stormwater (typical Danish stormwater has a lead median content around 30 μg/l). The lead was removed down to an average concentration of less than the detection limit (0.5 μg/l), corresponding to an overall removal rate of better than 97% (Figure 6). 22/87

23 Lead [mg m -3 ] 0 0, After sorption Figure 6. Removal of lead at the facility in Odense. The grey line illustrates the average values of the measurements Cadmium The concentration of cadmium in the inlet to the facility in Odense corresponded to typical Danish stormwater (typical Danish stormwater has a cadmium content of μg/l). The cadmium was removed down to an average concentration of less than the detection limit (0.05 μg/l), corresponding to an overall removal rate of better than 58% (Figure 7). Cadmium [mg m -3 ] 0, 0,0 After sorption Figure 7. Removal of cadmium at the facility in Odense. The grey line illustrates the average values of the measurements Chromium The concentration of chromium in the inlet to the facility in Odense corresponded to typical Danish stormwater (typical Danish stormwater has a chromium content of -70 μg/l).the chromium was removed down to an average concentration of less than the detection limit (0.5 μg/l), corresponding to an overall removal rate of better than 89% (Figure 8). 23/87

24 00 Chromium [mg m -3 ] 0 0, After sorption Figure 8. Removal of chromium at the facility in Odense. The grey line illustrates the average values of the measurements Nickel The concentration of nickel in the inlet to the facility in Odense corresponded to typical Danish stormwater (typical Danish stormwater has a nickel content of -90 μg/l). The nickel was removed down to an average concentration of 5. μg/l, corresponding to an overall removal rate of 72% (Figure 9). 000 Nickel [mg m -3 ] , After sorption Figure 9. Removal of nickel at the facility in Odense. The grey line illustrates the average values of the measurements Mercury The concentration of mercury in the inlet to the facilities was for nearly all analysis below the detection limit of 0.05 μg/l (typical Danish stormwater has a mercury content below.2 μg/l, and often below 0.05 μg/l). The mean concentrations in both the pond, after the sand filters and after the sorption filters were close to or beneath the detection limit (Figure 0). No conclusions on the effect of the sorption filters on mercury concentrations can hence be drawn. 24/87

25 Mercury [mg m -3 ] 0, 0,0 After sorption Figure 0. Removal of mercury at the facility in Odense. The grey line illustrates the average values of the measurements Total Nitrogen The concentration of total nitrogen in the inlet to the facility in Odense corresponded to typical Danish stormwater (typical Danish stormwater has a total nitrogen median content around 2 mg/l). The total nitrogen was removed down to an average concentration of.0 mg/l, corresponding to an overall removal rate of 63% (Figure ). 00 Total nitrogen [g m -3 ] 0 0, After sorption Figure. Removal of total nitrogen at the facility in Odense. The grey line illustrates the average values of the measurements Total Phosphorous The concentration of total phosphorous in the inlet to the facility in Odense corresponded to typical Danish stormwater (typical Danish stormwater has a phosphorous median content around mg/l). The total phosphorous was removed down to an average concentration of mg/l, corresponding to an overall removal rate of 9% (Figure 2). 25/87

26 0 Total phosphorous [g m -3 ] 0, 0,0 After sorption Figure 2. Removal of total phosphorous at the facility in Odense. The grey line illustrates the average values of the measurements Orthophosphate The concentration of orthophosphate in the inlet to the facility in Odense corresponded to typical Danish stormwater (typical Danish stormwater has an orthophosphate-p median content around mg/l). The orthophosphate-p was removed down to an average concentration of mg/l, corresponding to an overall removal rate of 95% (Figure 3). Orthophosphate-P [g m -3 ] 0, 0,0 0,00 After sorption Figure 3. Removal of orthophosphate at the facility in Odense. The grey line illustrates the average values of the measurements Total Suspended Solids (TSS) The concentration of suspended solids in the inlet to the facility in Odense corresponded to typical Danish stormwater (typical Danish stormwater has a suspended solids median content around mg/l). The suspended solids were removed down to an average concentration of 4.0 mg/l, corresponding to an overall removal rate of 90% (Figure 4). 26/87

27 000 Total suspended solids [g m -3 ] , After sorption Figure 4. Removal of suspended solids at the facility in Odense. The grey line illustrates the average values of the measurements Oil and grease The concentration of oil and grease in the inlet to the facility in Odense was below 5 mg/l. The oil and grease was removed down to a median concentration below the detection limit (0. mg/l), corresponding to an overall removal rate of 86% (Figure 5). 0 Oil and grease [g m -3 ] 0, 0,0 After sorption Figure 5. Removal of oil and grease at the facility in Odense. The grey line illustrates the average values of the measurements 6-PAH after USEPA The concentration of PAHs measured as 6-PAH after USEPA in the inlet to the facility in Odense was comparable to typical stormwater, but in most cases total PAHs were below the detection limit of 0.0 μg/l (typical Danish stormwater has a total PAH median content of μg/l). After the sand filters and after the sorption filters, nearly all measurements were below the detection limit (Figure 6). No conclusions on the effect of the sorption filters on PAH concentrations can hence be drawn. 27/87

28 0 6-PAH (USEPA) [mg m -3 ] 0, 0,0 0,00 After sorption Figure 6. Removal of 6-PAH after USEPA at the facility in Odense. The grey line illustrates the average values of the measurements Discussion of the effect of fixed media sorption filters The sorption filters effectively reduced the concentrations of copper, zinc, phosphorous and total solids (Table 7). The filters furthermore ensured that the extremely high concentrations of copper due to the illicit discharges came well below the water quality criterion for Danish fresh waters. In addition to polish the water quality of the stormwater runoff, the filters turned out to be an effective protection against pollutants which would otherwise have caused immediate toxic effects on the aquatic environment. Table 7 Average pollutant reduction by the sorption filters at the facility in Odense After sand filter After sorption filter Reduction Lead [mg m-3] % Cadmium [mg m -3 ] 0.05 < Chromium [mg m -3 ] < Copper [mg m -3 ] % Mercury [mg m -3 ] 0.06 < Nickel [mg m -3 ] 6 5 7% Zinc [mg m -3 ] % 6PAH [mg m -3 ] % TSS [g m -3 ] % Total N [g m -3 ] % Total P [g m -3 ] % Orthophosphate P [g m -3 ] % Oil and fat [g m -3 ] % COD [g m -3 ] % Removal of pollutants at the facility in Århus The facility for testing iron enrichment of bottom sediments (Århus) was taken into operation in January On April 2, 2009, a total of 3,000 kg of iron chloride/sulfate solution (PIX 8 from Kemira Water Wium-Andersen T, Nielsen A H, Hvitved-Jacobsen T, Vollertsen J (in press). Reduction of stormwater runoff toxicity by wet detention pond. In press for Highway and Urban Environment. Book Series: Alliance for Global Sustainability Series 28/87

29 Danmark A/S) was added to the pond. The product contained 6 g Fe 3+ kg - in an acidic solution. The product was mixed with pond water and distributed on the water surface. For this purpose, water was pumped from the pond and mixed with product pumped from the product container. The mixture was then delivered to a pipe floating on the pond surface and repeatedly dragged across the surface. Hereby a fairly uniform distribution of the iron salt into the pond water and subsequently on the pond bottom was achieved. Copper The concentration of copper in the inlet to the facility in Århus corresponded to typical Danish stormwater (typical Danish stormwater has a copper median content around 20 μg/l). Prior to iron enrichment of bottom sediments copper was removed down to an average concentration of.8 μg/l, corresponding to an overall removal rate of 9%. After the enrichment, copper was removed down to an average concentration of 2.7 μg/l, corresponding to an overall removal rate of 89% (Figure 7) Copper [mg m -3 ] 0 0 0, 0, Figure 7. Removal of copper at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Zinc The concentration of zinc in the inlet to the facility in Århus corresponded to typical Danish stormwater (typical Danish stormwater has a zinc median content around μg/l). Prior to iron enrichment of bottom sediments zinc was removed down to an average concentration of 20.4 μg/l, corresponding to an overall removal rate of 90%. After the enrichment, zinc was removed down to an average concentration of 7. μg/l, corresponding to an overall removal rate of 95% (Figure 8). 29/87

30 Zinc [mg m -3 ] Figure 8. Removal of zinc at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Lead The concentration of lead in the inlet to the facility in Århus was low compared to typical Danish stormwater (typical Danish stormwater has a lead median content around 30 μg/l). Prior to iron enrichment of bottom sediments lead was removed down to an average concentration of 0.3 μg/l, corresponding to an overall removal rate of 94%. After the enrichment, lead was removed down to an average concentration of 0.6 μg/l, corresponding to an overall removal rate of 90% (Figure 9) Lead [mg m -3 ] 0 0 0, 0, Figure 9. Removal of lead at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Cadmium The concentration of cadmium in the inlet to the facility in Århus was for most samples below the detection limit of 0.05 μg/l (typical Danish stormwater has a cadmium content of μg/l). The mean concentrations in both the pond and after the sand filters were below the detection limit both before and 30/87

31 after the iron enrichment of the bottom sediments (Figure 20). No conclusions on the effect of the iron enrichment on cadmium concentrations can hence be drawn. Cadmium [mg m -3 ] 0, 0, 0,0 0,0 Figure 20. Removal of cadmium at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Chromium The concentration of chromium in the inlet to the facility in Århus corresponded to typical Danish stormwater (typical Danish stormwater has a chromium content of -70 μg/l). Prior to iron enrichment of bottom sediments chromium was removed down to an average concentration of 0.3 μg/l (i.e. below the detection limit of 0.5 μg/l), corresponding to an overall removal rate of 93%. After the enrichment, chromium was removed down to an average concentration of.6 μg/l, corresponding to an overall removal rate of 73% (Figure 2). However, most the measurements in the basin and after the sand filters were below the detection limit of 0.5 μg/l. No conclusions on the effect of the iron enrichment on chromium concentrations can hence be drawn. 3/87

32 00 00 Chromium [mg m -3 ] 0 0 0, 0, Figure 2. Removal of chromium at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Nickel The concentration of nickel in the inlet to the facility in Århus corresponded to typical Danish stormwater (typical Danish stormwater has a nickel content of -90 μg/l). Nickel was significantly higher in the pond compared to the inlet. The phenomenon was seen both before and after the iron enrichment of the bottom sediments. However, after the enrichment the nickel concentrations in the pond were still higher than before the addition. The reason for nickel in the pond being higher than nickel in the inlet to the basin is not known, but it cannot originate from the iron used for enrichment, as this had a far too low nickel content to allow for such increase. Most likely the increased nickel content was caused by nickel containing clay minerals in the soil in which the pond is placed. This theory is strengthened by the fact that the sand filter removed most of it, i.e. the nickel has been associated with suspended particles (Figure 22) Nickel [mg m -3 ] 0 0 0, Figure 22. Removal of nickel at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements 32/87

33 Mercury The concentration of mercury in the inlet to the facility in Århus was for nearly all analysis below the detection limit of 0.05 μg/l (typical Danish stormwater has a mercury content below.2 μg/l, and often below 0.05 μg/l). The mean concentrations in both the pond and after the sand filters were close to the detection limit both before and after the iron enrichment of the bottom sediments (Figure 23). No conclusions on the effect of the iron enrichment on mercury concentrations can hence be drawn. 0 0 Mercury [mg m -3 ] 0, 0, 0,0 0,0 Figure 23. Removal of mercury at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Total Nitrogen The concentration of total nitrogen in the inlet to the facility in Århus corresponded to typical Danish stormwater (typical Danish stormwater has a total nitrogen median content around 2 mg/l). Prior to iron enrichment, total nitrogen was removed down to an average concentration of 0.65 mg/l, corresponding to an overall removal rate of 66%. After the enrichment, total nitrogen was removed down to an average concentration of 0.72 mg/l, corresponding to an overall removal rate of 73% (Figure 24). 33/87

34 0 0 Total nitrogen [g m -3 ] 0, 0, Figure 24. Removal of total nitrogen at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Total Phosphorous The concentration of total phosphorous in the inlet to the facility in Århus corresponded to typical Danish stormwater (typical Danish stormwater has a phosphorous median content around mg/l). Prior to iron enrichment of bottom sediments total phosphorous was removed down to an average concentration of 0.7 mg/l, corresponding to an overall removal rate of 34%. After the enrichment, total phosphorous was removed down to an average concentration of 0.08 mg/l, corresponding to an overall removal rate of 73% (Figure 25). The removal before by the pond was better than after the sand filters, i.e. the sand filters released phosphorous. The release seemed to decrease with time as the sand was washed clean. Total phosphorous [g m -3 ] 0, 0, 0,0 0,0 Figure 25. Removal of total phosphorous at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements 34/87

35 Orthophosphate The concentration of orthophosphate in the inlet to the facility in Århus corresponded to typical Danish stormwater (typical Danish stormwater has an orthophosphate P median content around mg/l). Prior to iron enrichment of bottom sediments orthophosphate P was removed down to an average concentration of 0.09 mg/l, corresponding to an overall removal rate of 8%. After the enrichment, orthophosphate P was removed down to an average concentration of 0.06 mg/l, corresponding to an overall removal rate of 90% (Figure 26). The removal before by the pond was better than after the sand filters. The release seemed to decrease with time as the sand was washed clean. Orthophosphate P [g m -3 ] 0, 0,0 0, 0,0 0,00 0,00 Figure 26. Removal of orthophosphate at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Total suspended solids (TSS) The concentration of total suspended solids in the inlet to the facility in Århus corresponded to typical Danish stormwater (typical Danish stormwater has a suspended solids median content around mg/l). Prior to iron enrichment of bottom sediments TSS was removed down to an average concentration of 5.9 mg/l, corresponding to an overall removal rate of 89%. After the enrichment, TSS was removed down to an average concentration of 3.8 mg/l, corresponding to an overall removal rate of 94% (Figure 27). 35/87

36 Total suspended solids [g m -3 ] Figure 27. Removal of total suspended solids at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Oil and grease The concentration of oil and grease in the inlet to the facility in Århus corresponded to typical Danish stormwater. Prior to iron enrichment of bottom sediments oil and grease was removed down to an average concentration of 0.8 mg/l, corresponding to an overall removal rate of 86%. After the enrichment, oil and grease was removed down to an average concentration of 0.2 mg/l, corresponding to an overall removal rate of 88% (Figure 28). 0 0 Oil and grease [g m -3 ] 0, 0, 0,0 0,0 Figure 28. Removal of oil and grease at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements 6-PAH after USEPA The concentration of PAHs measured as 6-PAH after USEPA in the inlet to the facility in Århus was comparable to typical stormwater, but in most cases total PAHs were below the detection limit of 0.0 μg/l 36/87

37 (typical Danish stormwater has a total PAH median content of μg/l) (Figure 29). No conclusions on the effect of the iron enrichment on the PAH concentrations can hence be drawn PAH (USEPA) [mg m -3 ] 0, 0,0 0, 0,0 0,00 0,00 Figure 29. Removal of 6-PAH after USEPA at the facility in Århus. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Discussion of the effect of iron enrichment of bottom sediments The addition of the 3,000 kg of iron product to the wet pond in Århus took about 8 hours for 3 persons, i.e. it was rather labor-intensive. The added product was strongly acidic and caused the ph to drop below 4 for a few hours. However, some 0 hours after completion of the addition, the ph was up at 6.5 and after 2 weeks the ph was back at the same level as before the addition (around 8). During the addition the water became rust-red. However, the color disappeared within hours as the iron flocculated and precipitated. The addition caused algae and other particulates to flocculate and precipitate with the iron, and the water became visibly clearer. The iron enrichment of the bottom sediments did not cause any measurable reduction in chemical water quality parameters. Actually, for a number of heavy metals the pond water concentrations increased compared to the time before the addition and the treatment efficiency of the pond itself deteriorated correspondingly. The product used for iron enrichment of the bottom sediments could, however, not have been the direct cause of the reduced treatment efficiency, as the product contained much less of the problematic metals than could be accounted for by the increased concentrations in the pond water. After the iron addition, nickel concentration in the pond doubled to 54 mg m -3, i.e. around 8 times the inlet concentration. Other metals stayed consistently below the inlet concentrations. In principle, the addition could have caused the solution of bound heavy metals as the ph decreased. However, the conditions were only acidic for a few hours, and it seems therefore unlikely that the addition was the cause of the increased concentrations of heavy metals. The treated water exiting the sand filters had similar heavy metals concentrations before and after the addition of iron to the pond. I.e. the sand filter reduced the increased concentrations of the pond water to the same and constant level, indicating that the additional heavy metals present after the addition of iron to the pond were associated with suspended particles. 37/87

38 Even though the iron enrichment of the bottom sediments did not result in decreased concentrations of pollutants, it did counteract growth of algae in the pond (Figure 43). Compared to the pond in Odense, the spring algae bloom was delayed and very much shortened and the chlorophyll content was low the rest of the summer. Removal of pollutants at the facility in Silkeborg The facility for testing addition of aluminum to the stormwater (Silkeborg) was taken into operation in June The addition was achieved by flow proportional dosing of aluminum hydroxides/oxides to the incoming water (alumin_0 from Remondis Production GmbH). The product contained 55 g Al 3+ kg - in an alkaline solution. For every 0 m 3 of inflow, a preset amount of the product was added to the inlet pipe of the pond. The addition of aluminum started on April 23, 2009 and varying amounts were added in the following period. During the first month of addition, a rather high concentration was added to the inflowing stormwater, namely a total of 7 kg Al to a total of 370 m 3 of stormwater i.e. in a concentration of 46 g Al m -3. This dose was then reduced to 5.2 g Al m -3 which was added from May 5 to September 2, during which the total inflow was 6,820 m 3. The pond was subdivided into 3 compartments by earthen barriers in level with the permanent water surface. Copper The concentration of copper in the inlet to the facility in Silkeborg corresponded to typical Danish stormwater (typical Danish stormwater has a copper median content around 20 μg/l). Prior to aluminum addition copper was removed down to an average concentration of 5.0 μg/l, corresponding to an overall removal rate of 60%. After the addition had commenced, copper was removed down to an average concentration of 3.6 μg/l, corresponding to an overall removal rate of 78% (Figure 30) Copper [mg m -3 ] 0 0 0, 0, Figure 30. Removal of copper at the facility in Silkeborg. The left hand figure illustrates the removal prior to aluminum addition and the right hand figure illustrates removal after the addition commenced. The grey line illustrates the average values of the measurements Zinc The concentration of zinc in the inlet to the facility in Silkeborg corresponded to typical Danish stormwater (typical Danish stormwater has a zinc median content around μg/l). Prior to aluminum addition zinc was removed down to an average concentration of 20 μg/l, corresponding to an overall removal rate of 84%. After the addition had commenced, zinc was removed down to an average concentration of 2 μg/l, corresponding to an overall removal rate of 78% (Figure 3). 38/87

39 Zinc [mg m -3 ] Figure 3. Removal of zinc at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Lead The concentration of lead in the inlet to the facility in Silkeborg was low compared to typical Danish stormwater (typical Danish stormwater has a lead median content around 30 μg/l). Prior to aluminum addition lead was removed down to an average concentration of 0.6 μg/l, corresponding to an overall removal rate of 8%. After the addition had commenced, lead was removed down to an average concentration of. μg/l, corresponding to an overall removal rate of 87% (Figure 32) Lead [mg m -3 ] 0 0 0, 0, Figure 32. Removal of lead at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Cadmium The concentration of cadmium in the inlet to the facility in Silkeborg was for most samples below the detection limit of 0.05 μg/l (typical Danish stormwater has a cadmium content of μg/l). All but one measured concentrations in the pond and after the sand filters were below the detection limit both before 39/87

40 and after the aluminum addition (Figure 33). No conclusions on the effect of the aluminum addition on cadmium concentrations can hence be drawn. Cadmium [mg m -3 ] 0, 0, 0,0 0,0 Figure 33. Removal of cadmium at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Chromium The concentration of chromium in the inlet to the facility in Silkeborg corresponded to typical Danish stormwater (typical Danish stormwater has a chromium content of -70 μg/l). Prior to aluminum addition chromium was removed down to an average concentration of. μg/l, corresponding to an overall removal rate of 43%. After the aluminum addition commenced, chromium was removed down to an average concentration of.0 μg/l, corresponding to an overall removal rate of 73% (Figure 34). However, many measurements in the basin and after the sand filters were below the detection limit of 0.5 μg/l. No conclusions on the effect of the aluminum addition on chromium concentrations can hence be drawn. 0 0 Chromium [mg m -3 ] 0, 0, Figure 34. Removal of chromium at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements 40/87

41 Nickel The concentration of nickel in the inlet to the facility in Silkeborg corresponded to typical Danish stormwater (typical Danish stormwater has a nickel content of -90 μg/l). Nickel decreased in the pond but increased significantly after the sand filter (Figure 35). I.e. the sand filters released nickel. The release seemed to decrease with time as the sand was washed clean Nickel [mg m -3 ] , 0, Figure 35. Removal of nickel at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Mercury The concentration of mercury in the inlet to the facility in Silkeborg was for the majority of measurements below the detection limit of 0.05 μg/l (typical Danish stormwater has a mercury content below.2 μg/l, and often below 0.05 μg/l). The mean concentrations in both the pond and after the sand filters were close to the detection limit both before and after the aluminum addition (Figure 36). No conclusions on the effect of the aluminum addition on mercury concentrations can hence be drawn. 4/87

42 00 00 Mercury [mg m -3 ] 0 0, 0 0, 0,0 0,0 Figure 36. Removal of mercury at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Total Nitrogen The concentration of total nitrogen in the inlet to the facility in Silkeborg corresponded to typical Danish stormwater (typical Danish stormwater has a total nitrogen median content around 2 mg/l). Prior to aluminum addition total nitrogen was removed down to an average concentration of 0.58 mg/l, corresponding to an overall removal rate of 76%. After the aluminum addition, total nitrogen was removed down to an average concentration of 0.38 mg/l, corresponding to an overall removal rate of 66% (Figure 37) Total nitrogen [g m -3 ] 0 0 0, 0, Figure 37. Removal of total nitrogen at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Total Phosphorous The concentration of total phosphorous in the inlet to the facility in Silkeborg was low compared to typical Danish stormwater (typical Danish stormwater has a phosphorous median content around mg/l). 42/87

43 Prior to aluminum addition total phosphorous was removed down to an average concentration of 0.09 mg/l, corresponding to an overall removal rate of 89%. After the aluminum addition, total phosphorous was removed down to an average concentration of 0.08 mg/l, corresponding to an overall removal rate of 85% (Figure 38). 0 0 Total phosphorous [g m -3 ] 0, 0,0 0, 0,0 0,00 0,00 Figure 38. Removal of total phosphorous at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Orthophosphate The concentration of orthophosphate in the inlet to the facility in Silkeborg was low compared to typical Danish stormwater (typical Danish stormwater has a orthophosphate P median content around mg/l). Prior to aluminum addition orthophosphate P was removed down to an average concentration of mg/l, corresponding to an overall removal rate of 95%. After the aluminum addition, orthophosphate P was removed down to an average concentration of mg/l (Figure 39). The inlet orthophosphate measurements after aluminum addition had commenced were very low. The most likely reason is that added aluminum has cross-contaminated the sampling point, causing complexation of the phosphate. 43/87

44 Orthophosphate P [g m -3 ] 0, 0,0 0, 0,0 0,00 0,00 Figure 39. Removal of orthophosphate at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Total suspended solids (TSS) The concentration of total suspended solids in the inlet to the facility in Silkeborg corresponded to typical Danish stormwater (typical Danish stormwater has a suspended solids median content around mg/l). Prior to aluminum addition TSS was removed down to an average concentration of 2.8 mg/l, corresponding to an overall removal rate of 84%. After the aluminum addition, TSS was removed down to an average concentration of.0 mg/l, corresponding to an overall removal rate of 98% (Figure 40) Total suspended solids [g m -3 ] 0 0 0, 0, Figure 40. Removal of total suspended solids at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Oil and grease Due to small sample volumes from the inlet, the inlet measurements were rather limited. Oil and grease in the pond and after the filters were comparable to the two other facilities (Figure 4). 44/87

45 0 0 Oil and grease [g m -3 ] 0, 0, 0,0 0,0 Figure 4. Removal of oil and grease at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements 6-PAH after USEPA Due to small sample volumes from the inlet, the inlet measurements were rather limited. 6-PAH after USEPA in the pond and after the filters were comparable to the two other facilities (Figure 42). 0 6-PAH (USEPA) [g m -3 ] 0, 0,0 0, 0,0 0,00 0,00 Figure 42. Removal of 6-PAH after USEPA at the facility in Silkeborg. The left hand figure illustrates the removal prior to iron enrichment of bottom sediments and the right hand figure illustrates removal after the enrichment. The grey line illustrates the average values of the measurements Discussion of the effect of aluminum addition The addition of aluminum to the stormwater runoff entering the facility in Silkeborg was much less labor intensive than the iron enrichment of the bottom sediments of the facility in Århus. The work was restricted to calibrating the dosing to the desired dosage level. As is seen from Figure 43, the dosing of 45/87

46 aluminum was very effective to hinder algae blooms in the pond more so than the iron enrichment of bottom sediments in Århus. However, when it came to chemical water quality parameters, no reduction could be seen. Actually, there was a slight increase in most heavy metal concentrations in the pond compared to before the addition. The difference in heavy metal concentrations from before the addition to during the addition was, however, not statistically significant. After passing the sand filters the water concentrations were similar to those before the aluminum addition. Sediment Sedimentation is considered the most important process in stormwater runoff pollution removal. Surface water runoff from rain and storm events comes into contact with particulate and soluble pollutants causing the runoff to incorporate them and become contaminated. Without effective treatment, these pollutants are transported via the runoff into receiving waters. Effective stormwater treatment immobilizes particles by depositing them on the bottom of the pond. The pollutant type and load will depend mainly on the pluviometric regime, the physicals characteristics of the catchment and the size and specific weight of the particles. The design of the three systems has taken sedimentation into account and the systems are dimensioned to enhance the opportunity of the incoming particles to settle. The accretion of sediment in a stormwater treatment system is proportional to the flow and the total suspended solids entering the system. The expected accumulation of sediment in the three systems will depend on the incoming total suspended solids (TSS) and, as mentioned before in the pollutant removal section, the load of TSS for the three systems is typical of stormwater treatment systems. Additionally, the average removal of TSS in the three systems is at least 90% showing that the design is adequate and that sediment will accumulate as the systems age. A supplementary source of sediment is the death and decay of biological material growing in the water that settle out and increase the amount of material deposited on the bottom of the ponds. To evaluate the sediment accretion and the pollutants in the systems, the depth of the accumulated sediment was measured by taking cores from the bottom in transects in three different parts of the ponds (inlet, middle and outlet) where possible. Samples for analysis of pollutants were taken by scraping the upper layer of accumulated sediment in the ponds at the same locations as the cores. Sediment samples analyzed for dry and organic content, nutrients and heavy metal concentration. The elements analyzed included phosphorus, potassium, calcium, iron, sodium, manganese, and aluminum. The heavy metals analyzed were lead, zinc, cadmium, nickel, chromium and copper. Additionally, the sediment was also analysed for6-pah recommended by USEPA. For some of the heavy metals, the concentrations in the sediment were below the detection limits (see Table 8 ). When a measurement was below the detection limit, half of the detection limit was assumed in the calculations. 46/87

47 Table 8. Detection limits for element analysis of sediments by ICP-OES (per dry weight basis). Parameter Unit Detection limit Phosphorus (P) mg/g d.w. 0.0 Potassium (K) mg/g d.w Calcium (Ca) mg/g d.w Sodium (Na) mg/g d.w Iron (Fe) mg/g d.w. 0.0 Manganese (Mn) mg/g d.w Aluminum (Al) mg/g d.w Lead (Pb) µg/g d.w. 2 Zink (Zn) µg/g d.w. 2 Cadmium (Cd) µg/g d.w. 0.5 Nickel (Ni) µg/g d.w. 0.5 Chromium (Cr) µg/g d.w. 0.5 Copper (Cu) µg/g d.w. 5 Nutrients and heavy metals in sediment Nutrients and heavy metals are typically found in soluble as well as attached to the particles in urban stromwater runoff. Heavy metals such as lead, chromium, zinc and copper are commonly found in urban environments originating from roofs, automobiles, paints and city infrastructure. The pollutants are transported, either in solution or attached to the suspended solids and can affect receiving waters, since they can be toxic to life or can accumulate in aquatic animals. The rate at which the particles accumulate in the wetponds will depend among others on the concentration of suspended solids coming to the systems; the characteristics of the catchment and the flow (see water quality section). Odense Sediment from the system in Odense was collected in two campaigns, December 2008 and July /87

48 Table 9. Average concentration of nutrients and heavy metals in sediment collected from the facility in Odense. Parameter Unit Phosphorus (P) IN MID OUT IN MID OUT Avg* Std** Avg* Std** Avg* Std** Avg* Std** Avg* Std** Avg* Std** mg/g dw Potassium (K) mg/g dw Calcium (Ca) mg/g dw Sodium (Na) mg/g dw Iron (Fe) mg/g dw Manganese (Mn) mg/g dw Aluminum (Al) mg/g dw Lead (Pb) µg/g dw Zink (Zn) µg/g dw Cadmium (Cd) µg/g dw <0.5 - <0.5 - <0.5 - <0.5 - <0.5 - <0.5 - Nickel (Ni) µg/g dw Chromium (Cr) µg/g dw Copper (Cu) µg/g dw * Average ** Standard deviation Phosphorus (P) The concentrations of phosphorus entering the system is typical of stormwater runoff (Danish stormwater has a phosphorous median content around mg/l). According to the performance review the removal of phosphorus in the systems is effective and always higher than 90%, which means that is likely that a high amount of the incoming phosphorous is retained in the sediment. The first year (2008) the concentration of P in the sediment is relatively low and higher concentrations are found at the outlet. In 2009 the concentrations of P along the flow increased in all the sampling points compared to P measured in 2008 reaching concentrations close to.0 mg/g d.w. in the middle and outlet of the system. Potassium (K) Potassium in the sediment of the system was relatively constant along the flow and for both years with concentrations between 3.0 and 4.0 mg/g d.w. Calcium (Ca) Calcium concentrations in the sediment are similar for both years and along the flow around 30 mg/g d.w. Sodium (Na) Sodium in the sediment was higher during 2008 and decreased along the water flow being the lowest at the exit of the system. In 2009 the sodium concentration in the systems was similar in the three transect with concentration around 0.30 mg/g d.w. Iron (Fe ) Iron is an element commonly found in soil and sediments. The concentrations in the sediment collected from Odense are similar in both years as well as along the water flow, ranging from ca. 20 to 30 mg/g d.w. The concentrations are similar to sediments found in nature. 48/87

49 Manganese (Mn) In 2008, the concentration of manganese in the system increased by around 20% from ca mg/g d.w. at the inlet and mid section transects to ca 0.6 mg/g d.w. at the outlet. In 2009 the concentrations of manganese were consistently lower compared to the previous year and the concentrations were similar in the different section of the ponds. Aluminum (Al) Aluminum is a natural constituent of soils and occurs in high concentrations in clay. The average concentration of Al in the inlet water was about 5 times the Al found in typical stormwaters. In 2008 the concentrations measured in the collected sediments were constant along the water flow with around 35 mg/g d.w. In 2009 the concentrations in the sediment were lower, about half of the values registered in the previous year. Lead (Pb) Lead is a heavy metal commonly found in significant concentrations in stormwaters. The lead concentration in the inflow water was within the expected concentrations for typical stormwaters, and overall removal close to 00% was recorded, indicating that the metal is expected to accumulate in the sediment. The concentrations in 2008 were higher at the inlet of the system and decreased along the water flow reaching /3 at the outlet transect (ca. 00 µg/g d.w.). For 2009 the same pattern was observed, but the concentrations were lower. However, the difference between the years was significant only for the samples from the middle of the pond. Zinc (Zn) Zinc is a heavy metal present in most stormwaters. The zinc concentrations in the incoming water for the system were considered to be in the higher average for stormwaters in Denmark. The system removes close to 00% of the zinc entering and therefore Zn is retained on the bottom of the pond. In 2008 the concentrations in the sediment were higher at the inlet and decreased significantly along the pond to about one half of the concentration at the inlet. In 2009, the concentrations in the sediment were lower, but the decreasing tendency along the length of the pond was maintained. The Zn concentration at the inlet was 000 µg/g d.w, while at the outlet of the system the concentrations was ca. 650 µg/g d.w. Cadmium (Cd) Cadmium concentration in the sediment for both years and all the transects was always below the detection limit (<0.5 µg/g d.w.). Nickel (Ni) Nickel concentration in the sediment of the system in Odense is similar for both years and decreasing significantly along the flow from ca. 40 µg/g d.w. down to ca 20 µg/g d.w. Chromium (Cr) In 2008 the Cr concentrations decreased along the system from ca. 00 µg/g d.w. to around 40 µg/g d.w. In 2009 the pattern was the same, but concentrations were lower, although not statistically significant.. Copper (Cu) The system in Odense is affected by high apparently uncontrolled discharges of copper. Consequently, the sediment is affected by the high loading of copper. In 2008, the Cu concentration in the sediments at the inlet of the wetpond was close to 4000 µg/g d.w. and the concentration decreased significantly along the pond to about half of the concentration (800 µg/g d.w.) In 2009, the same pattern was seen, but the Cu 49/87

50 concentrations were lower. At the inlet, the Cu concentration was ca µg/g d.w. and at the outlet end about 400 µg/g d.w. However, the statistical analyzes did not show significant differences in Cu concentrations between the years. Århus Sediment from the system in Århus was collected in three campaigns, November 2008, May 2009 and August The two sampling campaigns in 2009 were after the Fe dosing. Table 0. Average concentration of nutrients and heavy metals in sediment collected from the facility in Århus. Parameter Unit IN MID OUT IN MID OUT Ave* Std** Ave* Std** Ave* Std** Ave* Std** Ave* Std** Ave* Std** Phosphorus (P) mg/g dw Potassium (K) mg/g dw Calcium (Ca) mg/g dw Sodium (Na) mg/g dw Iron (Fe) mg/g dw Manganese (Mn) mg/g dw Aluminum (Al) mg/g dw Lead (Pb) µg/g dw Zink (Zn) µg/g dw Cadmium (Cd) µg/g dw <0.5 - <0.5 - <0.5 - <0.5 - <0.5 - <0.5 - Nickel (Ni) µg/g dw Chromium (Cr) µg/g dw Copper (Cu) µg/g dw * Average ** Standard deviation Phosphorus (P) Particulate phosphorous is a common contaminant in rain and stormwater runoff. The concentrations of phosphorus entering the Århus system is typical of stormwater runoff (Danish stormwater has a P median content around mg/l). The removal of P in the systems was effective and always higher than 90%, which means that P was retained in the wetpond sediment. The P concentrations in the sediment for both years was relatively constant throughout the pond, except for the outlet in 2009, where the concentration was around 20% higher than in the middle of the system. Potassium (K) Potassium concentrations in the sediment were generally in the range of 2.8 to 6.2 mg/g d.w. Calcium (Ca) Calcium is commonly found in urban stormwater in concentrations depending on the characteristics of the catchment. In 2008, the Ca concentration was around mg/g d.w. In 2009 the Ca concentrations were lower, 8-38 mg/g d.w. Sodium (Na) The Na concentrations in the sediments varied between 0.22 and 0.38 mg/g d.w. with no consistent differences between years and the different sections of the system. 50/87

51 Iron (Fe ) In 2008 and 2009 Fe concentrations are similar thorough out the pond with a concentration of ca 20 mg/g d.w. Apparently there was no effect of the iron applied, as no increase of iron was noticed in the sediment. Manganese (Mn) Manganese concentrations were constant in 2008 with concentrations around 0.40 mg/g d.w. In 2009 concentrations were slightly higher, but within the same range. Aluminum (Al) Aluminum concentration in 2008 were lowest at the inlet transects with a concentration of 26 mg/g d.w. and higher, mg/g d.w. in the mid and outlet sections. In 2009, the concentrations were lower, around 20 mg/g d.w. throughout the pond. Lead (Pb) Lead concentrations were 2-5 µg/g d.w. in 2008, but significantly higher, 9-24 µg/g d.w. in Zink (Zn) Zink concentration in 2008 were µg/g d.w.. In 2009, the Zn concentrations in the inlet and middle sections of the pond were nearly 50% higher compared to 2008, with concentration of µg/g d.w., Cadmium (Cd) Cadmium concentrations in the sediment for both years and all the transects were always below the detection limit (<0.5 µg/g d.w.).. Nickel (Ni) Nickel concentration in the sediment of the system in Århus was similar in 2008 and 2009 with concentrations in the three sections around 20 µg/g d.w. Chromium (Cr) Chromium concentrations in the sediments at Århus varied between 27 and 5 µg/g d.w. iwith no consistent pattern between years or sections of the system. Copper (Cu) Copper concentrations in the sediments were very different between the years 2008 and In 2008 the average Cu concentration in the sediment was around 20 µg/g d.w. while in 2009 the average concentrations of Cu from the same transects were to 280 µg/g d.w. 5/87

52 Silkeborg Sediment from the system in Silkeborg was collected in December 2008 and June Table. Average concentration of nutrients and heavy metals in metals in sediment collected from the facility in Silkeborg Parameter Unit Av e* IN MID OUT IN MID OUT Std** Ave* Std** Ave* Std** Ave* Std* * Ave* Std** Ave* Std** Phosphorus (P) mg/g dw Potassium (K) mg/g dw Calcium (Ca) mg/g dw Sodium (Na) mg/g dw Iron (Fe) mg/g dw Manganese (Mn) mg/g dw Aluminum (Al) mg/g dw Lead (Pb) µg/g dw 6 7 <2 - < <2 - <2 - Zink (Zn) µg/g dw Cadmium (Cd) µg/g dw < <0.5 - <0.5 - <0.5 - <0.5 - <0.5 - Nickel (Ni) µg/g dw Chromium (Cr) µg/g dw Copper (Cu) µg/g dw 7 0 <5 - < * Average ** Standard deviation Phosphorus (P) The concentrations of P in the water entering the system are lower than in typical stormwater runoff in Denmark. Hence, P concentrations in the sediment were also low. The P concentrations were significantly higher in the first section of the wetpond both years. Potassium (K) In both years K concentrations in the sediment decreased significantly along the system. In 2008, the K concentration decreased ca. 30% between the first and the second transect, as well as between the second and the last section. In 2009, the decrease was similar, although with slightly higher concentrations. Calcium (Ca) Calcium concentrations were generally low, and highest at the inlet pond. Sodium (Na) Sodium concentrations in the sediment decreased significantly in both years along the transects of the pond, and the concentrations of Na in 2009 were slightly higher. Iron (Fe ) Iron concentrations in the sediments were low and decreased significantly along the flow path both years. The Fe concentrations were higher in Manganese (Mn) Manganese showed a similar trend both years, with the highest concentrations measured at the inlet of the system. The concentrations in 2009 were lower than in 2008, except in the outlet pond. 52/87

53 Aluminum (Al) The concentration of Al in the sediment was similar both years around 30 mg/g d.w. at the inlet, decreasing to mg/g d.w at the outlet. Lead (Pb) Lead was only detected in the sediment of the first section of the system and with similar concentrations for both years of ca. 5 µg/g d.w. At the second and third sections, the Pb concentrations were below the detection limit (2 µg/g d.w.). Zink (Zn) Significantly higher Zn concentrations were found in the first section of the system, and in 2009 the concentrations were about 40% higher than in 2008 (472 as opposed to 285 µg/g d.w.). In the second and third section of the system, the Zn concentrations were much lower, µg/g d.w., with slightly higher concentrations in Cadmium (Cd) Cadmium concentration in the sediment for both years and in all the transects was always below the detection limit (<0.5 µg/g d.w.). Nickel (Ni) Nickel concentrations in the sediment were about 3 times higher at the inlet pond compared to the outlet pond both years, and the concentrations were generally higher in Chromium (Cr) Chromium concentrations in the sediment decreased by a factor of two from the inlet to the outlet of the pond. The concentrations generally increased with time. Copper (Cu) In 2008, Cu was only detected in the first section with a concentration of 7 µg/g d.w.. In 2009, the Cu concentration had increased to 72 µg/g d.w. in the first section, and also in the middle and outlet section, the Cu concentrations were now above the detection level (6-4 µg/g d.w.). Organic content Sediment samples were taken for the three systems at three transects along the water flow, namely inlet, middle and outlet of the systems. The organic content of the samples was analyzed as loss on ignition at 550ºC and expressed as % of dry weight. Table 2. Average organic content (% dry weight) of sediment samples collected from transects of the three systems. Section Odense Århus Silkeborg Ave* Stdv** Ave* Stdv** Ave* Stdv** Ave* Stdv** Ave* Stdv** Ave* Stdv** (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) ,4 6,7 7,6 2,4 6,3 2,7 2 3,6 Middle ,3 3,5 9,0 0,8 2,0 0,7 2,2 0,9 Out 8,7 2,8 9,0,8 5,3 2,2 7,9,3 0,9 0,4,0 0, * Average ** Standard deviation 53/87

54 The sediments in the system at Odense had a significantly higher organic content, particularly at the inlet end, compared to the sediments at Århus and Silkeborg. And in all systems, the organic content of the sediment decreased significantly from the inlet end to the outlet end. The content of organic matter in the sediments depends on the characteristics of the suspended in the the storm water. A high organic content is will be obtained when the TSS contains a significant amount of organic matter, whereas the organic content will be low when the TSS is mainly comprised of inorganic particles such as sand, silt and clay. Hence, the different organic contents in the sediments of the three systems are related to the catchment and the characteristics of the stormwater. PAHs in sediment The 6 PAHs recommended by USEPA were analysed in sediments collected from the three systems at transects at the inlet, middle section and outlet of each system. The PAHs were analysed by GC-MS after extraction of the sediments. Table 3. Median concentrations of the 6 PAHs after USEPA in sediment collected from inlet, middle and outlet sections of the three systems. Section Odense Århus Silkeborg Median Stdv Median Stdv Median Stdv (ng/g d.w.) (ng/g d.w.) (ng/g d.w.) (ng/g d.w.) (ng/g d.w.) (ng/g d.w) Middle Outlet of the PAHs recommended by USEPA where detected in the sediments from the 3 systems. The concentrations of PAHs were highest at Odense, and decreased with distance from the outlet in this systemthe PAH with the highest median concentration in the Odense pond was pyren (262, 272 and 4 ng/g d.w.), while acenaphthene was the only PAHs below the detection limit. In Århus, the sediment PAH concentrations were lower than in Odense, and the highest concentrations were found in sediments in the midsection of the system with 073 ng/g d.w. Of the 6 PAHs, only acenaphthene and fluoren were below the detection limits. The PAHs with the highest concentrations were pyren, benz(b)fluoranthen and indeno(,2,3-cd)pyren with similar concentration of ca. 80 ng/g d.w.. Antracen was found in low concentrations at the inlet of the pond (8 ng/g d.w.) and not detected in the middle and outlet section. In Silkeborg the total PAH concentrations in the sediment were lower than in the other two systems. The PAH with the highest concentration was pyren with ca 40 ng/g d.w. while acenaphthene and fluoren were below the detection limit in all samples. Most of the 6 priority PAHs were detected in the sediments of the three systems, and only acenaphthene was not detected in any of the systems. Fluoren was not detected in Århus and Silkeborg. The system in Odense had the highest concentrations of total PAHs with concentrations in the inlet sediment about 3 times as high as the concentrations in Århus, and about 7 times the concentrations in sediment at the inlet of the wetpond in Silkeborg. The difference in concentrations between the three systems is probably due to the location, the loadings of PAHs in the runoff water and the age of the systems. 54/87

55 Algae growth in the ponds Comparing the algae growth in the 3 ponds, both the iron enrichment and the aluminum addition were effective in reducing the algae growth. Most of the time the algae concentrations in the ponds, measured as chlorophyll concentration, of Århus and Silkeborg were a few percent of what was found in the Odense pond. This was the case even though no difference in the total phosphor and orthophosphate concentrations were observed. The reason here for is believed to be that the measurements only covered a growth season and not a winter season. During the growth season, the algae will take up most bio-available phosphorous. However, when adding iron or aluminum, the bio-available phosphorous fraction becomes severely limited, inhibiting the algae growth. Figure 43. Algae concentrations in the 3 ponds The positive effect of both iron and aluminum on keeping algae blooms within bonds was clearly visible in the ponds of Århus and Silkeborg, were the visual water quality was much better than in the pond of Odense. The removal effectiveness of the wet ponds The pollutant removal by the ponds themselves is shown in Table 4 and was within the range typically reported for such facilities 2. However, some anomalies were seen. The pond in Odense had comparatively poor treatment efficiency towards copper and zinc. The main part of these metals is typically associated with fine stormwater particles that settle out in wet ponds. However, in this case the metals were probably illicitly discharged to the stormwater system as dissolved or colloidal bound metals in rather high concentrations, resulting in poor removal efficiency. Another anomaly was seen in the pond in Århus, which consistently produced high concentrations of nickel while other metals were efficiently reduced. The 2 For example: Hvitved-Jacobsen T, Vollertsen J, Nielsen AH (200). Urban and Highway Stormwater Pollution: Concepts and Engineering. CRC Press, pp 544, ISBN: (Publication Date: March 26, 200) Vollertsen J, Åstebøl SO, Coward JE, Fageraas T, Nielsen AH, Hvitved-Jacobsen T (2009b). Performance and Modeling of a Highway Wet Detention Pond Designed for Cold Climate. Scheduled for publication Water Quality Research Journal of Canada, in 44(3) Hvitved-Jacobsen T, Vollertsen J, Nielsen AH (200). Urban and Highway Stormwater Pollution: Concepts and Engineering. CRC Press, pp 544, ISBN: (Publication Date: March 26, 200) 55/87

56 reason is not known, however, it seems likely that the nickel has been present in the natural soil minerals of the pond and slowly released to the bulk water. The facility in Silkeborg did also show a slight anomaly in terms of rather poor removal of zinc. The reason for this is not known but it could be the same as for nickel release at the facility in Århus. Table 4. Average pollutant reduction by the wet ponds. For The facility in Århus and Silkeborg, only data up till start of iron/aluminum addition are included Facility in Odense Facility in Århus Facility in Silkeborg Outlet Reduction Outlet Reduction Outlet Reduction Lead [mg m -3 ] % % % Cadmium [mg m -3 ] % 0.08 < < Chromium [mg m -3 ] % % % Copper [mg m -3 ] % % % Mercury [mg m -3 ] % 0.30 < < Nickel [mg m -3 ] % % % Zinc [mg m -3 ] % % % 6PAH [mg m -3 ] % % TSS [g m -3 ] % % % Total N [g m -3 ] % % % Total P [g m -3 ] % % % Ortho P [g m -3 ] % % 0.76 < Oil and fat [g m -3 ] %.0. -8% COD [g m -3 ] % % The removal effectiveness of the sand filters The concentration of most of the pollutants was efficiently reduced by the sand filters (Table 5). The sand filters were especially efficient for reduction of the rather high concentrations of copper, lead and zinc at the facility in Odense. However, phosphorous was not reduced by the sand filters. At the facility in Århus, the sand filters actually increased the phosphorous concentration in terms of both total phosphorous and orthophosphate. Similarly did the concentration of nickel at the facility in Silkeborg increase dramatically by passing the sand filters. At the facility in Århus and Odense, the sand filters furthermore released significant amounts of iron oxides, causing about 2 and 7 g m -3 iron in the outlet from the sand filters, respectively. At the facility in Århus the iron release from the sand filters was monitored closely and a clear decrease in iron release was observed over time. This observation indicates that the sand filters over a year or two probably will have washed clean of dissolvable compounds. The iron release was likely cause by part of the sand going anaerobic, causing iron reduction and release, which then was re-oxidized after exiting the filter. 56/87

57 Table 5. Average pollutant reduction by the sand filters. For The facility in Århus and Silkeborg, only data up till start of iron/aluminum addition are included Facility in Odense Facility in Århus Facility in Silkeborg Pond After sandfilter Reduction Pond After sandfilter Reduction Pond After sandfilter Reduction Lead [mg m -3 ] % % % Cadmium [mg m -3 ] % <0.05 < <0.05 < Chromium [mg m -3 ].2 < < % Copper [mg m -3 ] % % % Mercury [mg m -3 ] % < <0.05 < Nickel [mg m -3 ] % % % Zinc [mg m -3 ] % % % 6PAH [mg m -3 ] % % 0.05 <0.0 - TSS [g m -3 ] 8 4 8% 6 6-6% 3 3 8% Total N [g m -3 ] % % % Total P [g m -3 ] % % % Ortho P [g m -3 ] % % <0.005 < Oil and fat [g m -3 ] % % % COD [g m -3 ] % % % The hydraulic capacity of the sand filters was significantly lower than had been envisioned during design. When designing the filters, it was assumed that the comparatively large wet pond would cause less clogging of the filters compared to traditional infiltration basins and the filters were sized accordingly. It was furthermore assumed that the vertical filters would show little or no clogging, followed by the sloping filters and the horizontal filters clogging most. However, it turned out that all filters clogged more or less to the same extent and to a much higher degree than envisioned. The hydraulic capacity of the vertical sand filter and the sloping sand filter were most of the time below the detection limits of the flow meters. However, the outflow from the horizontal sand filter was measurable and was proportional to the water pressure on the filter surface (Figure 44). Even at still lower water levels than shown in Figure 3, the outflow continued to be proportional to the water pressure (data not shown). Assuming that the resistance to the flow lies in a thin clogging layer in the upper part of the sand filter, the flow out of the filter can be described by a simplified version of Darcy s law (Equation ). 57/87

58 Flow out of filter [L s - ] Water pressure on sand filter Flow out of sand filter 0,20 0,5 0,0 0, Water pressure on sand filter [m] ,00 Figure 44. The relative water pressure on the horizontal sand filter and the corresponding outflow at the facility in Århus Q L A h Eq. Where Q is the flow through the clogging layer [m 3 s - ], L is a leakage constant [s - ], A is the surface area and h is the water pressure at the clogging layer [m water column]. Analyzing the data shown in Figure 44 by means of Equation yields a leakage factor, L, of s -. Applying this parameter in Equation allows determination of the outflow at varying water depths and sand filter areas. E.g. for the sand filter of the facility in Århus from where the data originate, a water depth of 5 cm will give rise to.7 L s - of outflow, whereas a water depth of 50 cm will cause 7 L s - of outflow. On line measurement of water quality data The facility in Odense The ph, temperature, turbidity and oxygen concentration was measured in the pond and approximately half a meter below the water surface (Figure 45, Figure 46, Figure 47, Figure 48). In the period from July to August , the support for the measuring probes had tilted and the probes dug into the bottom sediments. The data in this period are therefore not trustworthy. The ph varied between 6.4 and above 0, with typical values between 7 and 9. The ph measurements from middle of March to middle of May 2009 are uncertain as the sensor drift in this period was large, causing a replacement of the ph sensor. The oxygen concentration varied from zero to above 20 g m -3, with the highest variability during the spring and summer where photosynthesis produced and consumed oxygen. The zero turbidity measurements during the summer and autumn 2008 were caused by sensor malfunctioning. 58/87

59 Figure 45. ph measurements in the pond in Odense Figure 46. Temperature measurements in the pond in Odense 59/87

60 Figure 47. Dissolved oxygen measurements in the pond in Odense Figure 48. Turbidity measurements in the pond in Odense The facility in Århus The ph, temperature, turbidity and oxygen concentration was measured in the pond and approximately half a meter below the water surface (Figure 49, Figure 50, Figure 5, Figure 52). ph in the pond of Århus 60/87

61 varied between 6.7 and 9.6 with typical values between 7 and 9. The highest variability was in the summer period and caused by algae growth. Compared to the summer period before the iron enrichment of the bottom sediments, the average ph level after the enrichment was approximately ph unit lower (Figure 49). Similarly was the oxygen level of the summer period after the addition was lower than that of the summer period before the addition (Figure 5). For oxygen the difference was rather pronounced. The most likely explanation is a combination of reduced photosynthesis due to reduced algae biomass and the fact that the pond received an unknown amount of septic wastewater, causing increased oxygen depletion in the pond. Figure 49. ph measurements in the pond in Århus. The sharp peaks are due to censor calibration 6/87

62 Figure 50. Temperature measurements in the pond in Århus Figure 5. Dissolved oxygen measurements in the pond in Århus 62/87

63 Figure 52. Turbidity measurements in the pond in Århus The facility in Silkeborg The ph, temperature, turbidity and oxygen concentration was measured in the middle section and the last section of the pond and approximately half a meter below the water surface (Figure 53, Figure 54, Figure 55, Figure 56). The ph variability in Silkeborg tended to be lower after the addition of aluminum had been initiated (Figure 53). Furthermore, the oxygen concentration in July, August, and September2009 were comparatively low (Figure 55). Similar to the pond in Århus, this was likely caused by the reduced photosynthesis caused by the reduced algae concentrations in the pond. 63/87

64 Figure 53. ph measurements in the pond in Silkeborg. Measurements from the middle section are indicated with pond ; measurements from the last section are indicated with last pond Figure 54. Temperature measurements in the pond in Silkeborg. Measurements from the middle section are indicated with pond ; measurements from the last section are indicated with last pond 64/87

65 Figure 55. Dissolved oxygen measurements in the pond in Silkeborg. Measurements from the middle section are indicated with pond ; measurements from the last section are indicated with last pond Figure 56. Turbidity measurements in the pond in Silkeborg. Measurements from the middle section are indicated with pond ; measurements from the last section are indicated with last pond 65/87

66 The turbidity measurements in the Silkeborg pond were rather unstable, and subject to rapid but very short increases in signal. It is not known whether this was caused by signal interferences or by natural means, e.g. snails passing the surface of the sensors. The data are still valid when ignoring these unmotivated peaks. To illustrate this fact, a week of measurements is shown in Figure 57, illustrating the continuous and valid signal over-layered by unmotivated peaks of short duration. Figure 57. Turbidity measurements in the pond in Silkeborg, one week of measurement. Measurements from the middle section are indicated with pond ; measurements from the last section are indicated with last pond Hydraulic characteristics of the ponds The facility in Odense The flow to the facility varied over the year as can be seen from Figure 58 and Figure 59. Especially the period from August to November 2008 was a rather wet period. 66/87

67 Figure 58. Flow into the facility in Odense Figure 59. Accumulated flow to the facility in Odense 67/87

68 Figure 60. Water level in the facility in Odense Estimation of catchment area using flow measurement The measured flow was compared to the flow calculated by the hydraulic model. This was done in order to calibrate and validate the contributing catchment area as well as the model used to estimate the flow and initial design criteria. To describe the collection system a MOUSE model was used. The physical sewer data were well registered, i.e. lengths, inverts levels and materials. The catchment areas were estimated using detailed drawings of pipes and drainage on each plot and high resolution aerial photos (Figure 6). Figure 6. Catchment area (pink) and model (blue) for the facility in Odense 68/87

69 Rain events were collected using a RIMCO tipping bucket rain gauge located in the city area. The rain gauge are used by the Danish Water Pollution Control Committee and connected to a national database. The rain gauge located closest to the study area was used (2882, approx. 2.7 km to the east). The rain was applied to each manhole as runoff hydrographs taking runoff time, reduction factor and initial losses into account. All measured and calculated flow events from February 2008 to September 2009 are shown in Figure 62. The model describes the total inflow to the facility well (R 2 =0.87). The contributing catchment area is validated to around.74 reduced hectares. A reduction factor of 0.9 was used to take into account that not all rain water on the surface will end up in the sewer. For this area which is highly urbanized with a large degree of impervious area a factor of is advised for future reference. This factor was found to be independent of the intensity or volume of the rain. Usually a factor of 0.9 is used for smaller less intense rain events. Figure 62. Total measured and calculated inflow for the facility in Odense 6 rain events are illustrated in Figure 63. The model simulates the flow and system dynamics well for both smaller and larger rain events. On the 25 th of August 2009 a rain front coming from the east is causing a delayed peak flow in the measurement compared to the calculated values. 69/87

70 Figure 63. Calculated and measured flow for 6 rain events in Odense. The facility in Århus In the beginning of the measuring period, the water level of the pond in Århus was low, as was the inflow to the pond (Figure 64, Figure 65, Figure 66). These measurements are artifacts, as they were caused by closing off of the large flow meter. The flow meter had been damaged by flooding of the measuring structure holding the flow meter and had to be replaced. In this period, only the small flow meter allowed passage of stormwater into the pond, causing most of the flow to go into the overflow structure. 70/87

71 Figure 64. Flow into the facility in Århus Figure 65. Accumulated flow to the facility in Århus 7/87

72 Figure 66. Water level the facility in Århus Estimation of catchment area using flow measurement The measured flow was compared to the flow calculated by the hydraulic model. This was done in order to calibrate and validate the contributing catchment area as well as the model used to estimate the flow and initial design criteria. To describe the collection system a MOUSE model was used. The physical sewer data were well registered, i.e. lengths, inverts levels and materials. The catchment areas were creating using areas estimated in the waste water plan developed by the municipality. No detailed information was available on the local drainage within each plot in the area (Figure 67). 72/87

73 Figure 67. Catchment area (pink) and model (blue) for the facility in Århus Rain events were collected using a RIMCO tipping bucket rain gauge located in the city area. The rain gauge located closest to the study area was used (2236, approx. 2 km south). The rain was applied to each manhole as runoff hydrographs taking runoff time, reduction factor and initial losses into account. All measured and calibrated calculated flow events from June 2009 to September 2009 are shown in Figure 68. The model describes the total inflow to the facility well (R 2 =0.8). The calibration indicates that the contributing area should be in the range of 8-20 reduced hectares instead of This will effect the calculated detention time in the basin. 73/87

74 Figure 68. Total measured and calculated inflow for the facility in Århus 6 rain events are illustrated in Figure 69. The model simulates the flow and system dynamics well for both smaller and larger rain events. 74/87

75 Figure 69. Calculated and measured flow for 6 rain events in Århus The facility in Silkeborg The flow to the facility varied over the year as can be seen from Figure 70 and Figure 7. The pond in Silkeborg was affected by leaks in the clay membrane, causing temporary low water levels in the pond (Figure 72). Towards the end of the measuring period, the leakage diminished and the water level stabilized. 75/87

76 Figure 70. Flow into the facility in Silkeborg Figure 7. Accumulated flow to the facility in Silkeborg 76/87

77 48,25 48,2 48,5 48, 48, ,95 47,9 47,85 47,8 47,75 47,7 47,65 47,6 47, : : : : : : : : : :00 Water level, pond [m] Figure 72. Water level the facility in Silkeborg Estimation of catchment area using flow measurement The measured flow was compared to the flow calculated by the hydraulic model. This was done in order to calibrate and validate the contributing catchment area as well as the model used to estimate the flow and initial design criteria. To describe the collection system a MOUSE model was used. The physical sewer data were well registered, i.e. lengths, inverts levels and materials. The catchment areas were originally created using digital technical maps and high resolution aerial photos (Figure 73). Figure 73. Catchment area (pink) and model (blue) for the facility in Silkeborg Rain events were collected using a RIMCO tipping bucket rain gauge located in the city area. The rain gauge are used by the Danish Water Pollution Control Committee and connected to a national database. The rain 77/87

78 gauge located closest to the study area was used (2249, approx. 2 km to the east). The rain was applied to each manhole as runoff hydrographs taking runoff time, reduction factor and initial losses into account. All measured and calculated flow events from June 2009 to September 2009 are shown in Figure 74. The model describes the total inflow to the facility well (R 2 =0.86). The contributing catchment area was calibrated using spectral analysis and measured flow and was estimated to be 7. reduced hectares. Only smaller rain events with a rain depth below 0 mm were observed during the period and therefore a reduction factor of 0.8 was used. This area is less urbanized with a smaller degree of impervious area compared to the facility in Odense. Figure 74. Total measured and calculated inflow for the facility in Silkeborg 6 rain events are illustrated in Figure 75. The model simulates the flow and system dynamics well for both smaller and larger rain events. 78/87

79 Figure 75. Calculated and measured flow for 6 rain events in Silkeborg Mixing and flow pattern in the ponds Turbidity changed due to stormwater inflow, where an inflow event caused a rapid increase in turbidity in the pond. An example hereof is given in Figure 76 and Figure 77 for the facility in Odense. From the latter figure it is evident the delay time from initiation of the inflow till the turbidity was affected amounted to a few hours. Similar response times were found when looking at the response of temperature, ph and oxygen on an inflow event. I.e. the time from a significant inflow event began until the pond was completely mixed was in the order of hours and hence significantly shorter than the average water residence time in the pond. The later calculated from numerical simulation of the measured events. 79/87

80 Figure 76. Turbidity and inflow into the pond in Odense in the March 2009 Figure 77. Turbidity and inflow during the runoff event of March 9, 2009, Odense Similar behavior could be seen in the pond in Århus. However, in the pond in Silkeborg such behavior was not observed. This was probably caused by the compartmentalized design of the pond, causing the rapid mixing to be restricted to the first compartment. This was also indicated by visual inspection of the 3 80/87