The Science of the Total Environment 235 Ž1999. 363]365
Treatment of highway runoff: a study of three detention ponds Karin LundbergU , Maria Carling, Per Lindmark Swedish Geotechnical Institute, S-581 93 Linkoping, Sweden ¨
Abstract We have performed a monitoring study of three detention ponds in the south of Sweden, with the aim to clarify the treatment efficiency in the ponds. Q 1999 Elsevier Science B.V. All rights reserved. Keywords: Highway runoff; Detention ponds; Treatment efficiency
1. Extended abstract Three wet ponds in southern Sweden have been investigated concerning treatment efficiency for road runoff. The project was commissioned by the Swedish National Road Administration. Although different kinds of treatment methods have become increasingly common in recent years, studies of their efficiency have seldom been performed. In Sweden, wet detention ponds are the most commonly used method for the treatment of highway runoff. Knowledge from the studied ponds and a literature study in the field have
U
Corresponding author. Tel: q46-13-20-18-12; fax: q4613-20-19-09. E-mail address:
[email protected] ŽK. Lundberg.
resulted in a report with design advice ŽSwedish National Road Administration, 1998.. During 1995, two detention ponds were studied; one on the E6 European highway south of Falkenberg and one on the Rv 13 national trunk road west of Klippan. Due to extensive erosion and clay suspension in the pond in Klippan, sampling was interrupted after 1 year and the pond was replaced with a detention pond on the E22 European highway near Karlskrona. The pond in Falkenberg was studied during 1995 and 1996. The ponds were chosen with regard to several aspects: traffic intensity, year of construction Žat least 1 or 2 years of operation., design of the pond Žwell defined inlet and outlet. and the source of the water. Water samples were taken at the inlet and
0048-9697r99r$ - see front matter Q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 9 9 . 0 0 2 3 6 - 3
364 Table 1 Mean values and concentration intervals in the studied ponds compared to background values in surface water ŽSwedish Environmental Protection Agency, 1996. and a reference sample in River Vinan ˚ close to the pond in Falkenberg Žbased on unfiltrated samples. a Falkenberg 1995 and 1996 Ž5 samples at 7 occasions.
Klippan 1995 Ž2 samples at one occasion.
Karlskrona 1999 Ž4]5 samples at one occasion.
Inlet
Outlet
Inlet
Outlet
lInlet
Outlet
PH
6.5]7.5 7.3
6.8 Ž5.7]7.4.
7.3 Ž6.4]8.0.
7.8 Ž7.8]7.9.
7.8 Ž7.7]7.9.
6.6 Ž6.2]7.1.
6.6 Ž6.2]6.9.
El. Cond ŽmSrm.
2]10 21.6
62 Ž23]94.
75 Ž58]103.
43 Ž41]44.
43 Ž40]46.
16 Ž8]26.
62 Ž61]63.
0
46 Ž9]197.
10 Ž0]215.
80 Ž70]80.
80 Ž80]90.
20 Ž12]26.
6 Ž1]20.
32 Ž- 30]50.
33 Ž- 30]50.
30
- 30
- 30
- 30
41 Ž35]46.
tot-N Žmgrl.
0.30 7.1
2.6 Ž1.6]4.7.
2.0 Ž0.64]8.7.
4.8 Ž4.8]4.9.
5.4 Ž4.3]6.4.
3.5 Ž1.9]5.8.
1.2 Ž1]1.4.
Cl Žmgrl.
10]15c 25.5
111 Ž45]200.
141 Ž80]190.
49 Ž48]50.
50 Ž43]57.
27 Ž12]42.
158 Ž150]160.
Cd Žmgrl.
0.01 0.08
0.36 Ž0.14]0.84.
0.23 Ž- 0.05]0.86.
0.16 Ž0.12]0.19.
0.12 Ž0.11]0.13.
0.17 Ž0.13]0.21.
0.08 Ž0.05]0.15.
Cr Žmgrl.
0.4 1.5
3.0 Ž- 0.5]8.4.
2.2 Ž- 0.5]6.5.
1.3 Ž1.2]1.4.
1.5 Ž1.1]1.8.
1.9 Ž1.6]2.3.
1.1 Ž0.8]1.4.
Cu Žmgrl.
0.3 1.7
17 Ž2.2]98.
7 Ž1.4]24.
8.0 Ž6.6]9.4.
7.4 Ž6.7]8.2.
22 Ž18]25.
12 Ž11]15.
Ni Žmgrl.
1 2.2
17 Ž4.8]41.
14 Ž3.4]40.
4.5 Ž4.3]4.6.
4.2 Ž3.7]4.7.
2.0 Ž1.5]2.6.
3.4 Ž2.8]4.0.
Pb Žmgrl.
0.2 0.79
9.0 Ž0.21]76.
3.1 Ž0.4]27.9.
6.2 Ž5.1]7.4.
5.4 Ž5.0]5.7.
6.6 Ž5.8]7.7.
4.5 Ž3.5]6.1.
Zn Žmgrl.
1 9.6
167 Ž29]782.
27 Ž2.2]122.
24 Ž21]26.
27 Ž17]36.
446 Ž167]574.
124 Ž70]170.
SS Žmgrl. COD Žmgrl.
a
Cells with bold font indicate parameters of special concern. The reference sample in Vinan, ˚ close to the pond in Falkenberg, corresponds to one filtrated sample taken on 11 November 1995. c Chloride concentration in groundwater, median in soil wells and rock wells ŽAastrup et al., 1995.. b
K. Lundberg et al. r The Science of the Total En¨ ironment 235 (1999) 363]365
Background ŽSwedish Environmental Protection Agency, 1996. Vinan ˚ b Surface water
K. Lundberg et al. r The Science of the Total En¨ ironment 235 (1999) 363]365
outlet of the ponds in connection with heavy rain events at different times of the year during 1995]1996 and analysed for pH, conductivity, suspended solids ŽSS., COD, total nitrogen, chloride and metals Žboth filtered and unfiltered.. A few organic analyses were also performed. Water moss Ž Fontinalis antipyretica. has been used as a longterm monitor throughout the sampling period in both the inlet and outlet of the ponds since moss is able to accumulate contaminants. From the metal analyses during the first year on filtered and unfiltered samples, it was concluded that most of the metals were present in the particulate phase. The results from the ponds indicated that Cu and Pb exist to a greater extent as particles, but a greater part of Cd and Zn exists as dissolved substances. Table 1 shows mean values and concentration intervals based on analyses of inlet and outlet water from the investigated ponds. Large variations in the concentrations of SS and metals in highway runoff were observed, depending on the season in which the sample was taken. The magnitude of the variation was at least 10-fold, but for Pb it was 100-fold. Additionally, the length of the preceding dry period was also correlated to higher concentrations of SS and metals in the runoff. Generally, it is noted that the concentration of suspended solids, total nitrogen and most of the
365
metals were lower in the outlet than in the inlet, which indicates that pollutant removal took place. A comparison with background values in surface water shows that total-N, Cd, Cu, Pb and Zn exhibited the clearest increase in concentrations. The concentration of lead is considerably lower in all three ponds compared to concentrations presented earlier in the literature, which is a clear result of the changeover to unleaded fuel. For the Falkenberg pond, the main part of the total-N concentration could be attributed to the surrounding agricultural area. Higher concentrations than expected in the pond effluents suggest that the three ponds were failing to work as intended. Possible explanations are that the ponds were not correctly dimensioned or that sedimented sludge was mobilised. These results suggest that further improvements in detention pond design are needed. References Aastrup M, Thunholm B, Johnson J, Bertills U, Berntell A. The chemistry of ground water in Sweden. NV Rapport 4415. SGU, 1995 win Swedishx. Swedish Environmental Protection Agency. Contaminated soils. Guidelines to inventories and risk assessments. Preliminary version. January 1996. NV, SGU, ITM, IMM, 1996 win Swedishx. Swedish National Road Administration. Treatment of highway runoff. Dimensioning advice. VV Publ. 1998:009, 1998 win Swedishx.
