the use of rain gardens as biofilters in highways and … · rain gardens, also called bioretention...

The use of rain gardens as biofilters in

highways and the effect of highway

runoff water in Daphnia magna and

Pseudokirchneriella subcapitata

Violeta Mejias Sanz

Company Supervisors: Maja Halling and Anna Sporre

University supervisors: Åsa Arrhenius and Thomas Backhaus

Master Thesis 2017

Ecotoxicology Master Program.

Index

Abstract ......................................................................................................................................... 3

Acknowledgments ......................................................................................................................... 4

1. Introduction ........................................................................................................................... 5

1.1. Aim ................................................................................................................................ 6

2. Methods ................................................................................................................................. 6

2.1. Rain gardens .................................................................................................................. 6

2.2. Highway runoff ............................................................................................................. 8

2.3. Toxicity tests ................................................................................................................. 8

2.4. Chemical analyses ......................................................................................................... 9

2.5. Statistical analyses......................................................................................................... 9

3. Results and Discussion ........................................................................................................ 10

3.1. Daphnia magna ........................................................................................................... 10

3.2. Pseudokirchneriella subcapitata ................................................................................. 11

3.3. Chemical Analyses ...................................................................................................... 16

4. Conclusions ......................................................................................................................... 20

5. References ........................................................................................................................... 21

6. Appendices .......................................................................................................................... 23

6.1. Appendix 1 .................................................................................................................. 23

6.2. Appendix 2 .................................................................................................................. 24

6.3. Appendix 3 .................................................................................................................. 25

6.4. Appendix 4 .................................................................................................................. 26

6.5. Appendix 5 .................................................................................................................. 32

Abstract

Rain gardens are efficient biofilters that remove pollutants, reduce flooding, recharge ground

water and provide habitat for plants and wildlife. Highway runoff is known for being one of the

most important diffuse pollution sources and few studies have been performed on rain gardens

and highway runoff in Sweden. The project aim was to study the effect of filtering water from

several typess of rain gardens watered with highway runoff and the runoff itself, using Daphnia

magna and Pseudokirchneriella subcapitata as bioindicators. There were three types of rain

gardens tested: just soil; one individual of three perennials species: Potentilla atrosanguinea,

Armeria maritime Alba, and Sedum acre; and two individuals of the three perennials previously

mentioned. The highway runoff samples were taken at the E6 in Mölndal during April, May, June,

August and September in 2016. In order to measure any possible toxicity , an acute mobility

inhibition test in Daphnia magna and a growth inhibition test in Pseudokirchneriella subcapitata

were performed. The mobility of Daphnia magna was affected by the runoff itself in August and

September, but not for the filtered water through the rain gardens. An inhibition effect on the

growth of Pseudokirchneriella subcapitata by the runoff was observed in May, June and

September at different concentrations and time of exposure. Only filtered water from August

and September was tested in Pseudokerchneriella subcapitata where a positive growth was

observed in the short run and an inhibition of growth appeared after six days. According to the

chemical analyses, high levels of zinc and copper were found in the runoff and moderate high

concentration of chromium and lead. It can be concluded that rain gardens have proven to be

efficient biofilters preventing lethal impacts on Daphnia magna and Pseudokeirchneriella

subcapitata, even though pollutants were not completely removed.

Acknowledgments

I would like to thank Maja Halling and Anna Sporre from EnviroPlanning for the great support.

To Åsa Arrhenius and Thomas Backhaus from Gothenburg University for their understanding and

help. In special to my husband for his patience and help under the whole project and to my little

beautiful daughters that gave me the motivation and strength to keep up.

1. Introduction

Rain gardens, also called bioretention areas, are “shallow depressions in the landscape, with

absorbent, free draining soil and planted with vegetation that can withstand occasional

temporary flooding” (Bray et al., 2012; Dietz and Clausen, 2005; Hinman, 2013). Through rain

gardens the volume of rainwater running off that may wash oils, heavy metals and other

pollutants into drains is reduced and low-level pollution is treated (Bray et al., 2012). It is

becoming increasingly popular, being recommended as best management practice (BMP) to

treat stormwater runoff and reduce nonpoint source pollution from urban areas (Prince

George´s County, 1993, 1999; Dietz and Clausen, 2005). Raingardens also allows the water to

infiltrate, recharge aquifers and reduce peak flows. In Sweden, biofilters or rain gardens are

quite new and few tests have been performed, but it is becoming a strong alternative to prevent

contamination.

Highway runoff is known to be one of the most important diffuse pollution sources, that can

deteriorate the quality of the water and cause severe damage to aquatic ecosystems affecting

species abundance, diversity an d even increasing non-native species or pollution-tolerant taxa

(Spromberg et al., 2015; Valtanen et al., 2015; Malinowska et al., 2015). The main sources of

pollution in roads are abrasion of tires and road surfaces, wear of brake linings and moving

engine parts, corrosion of vehicle components, oil spills and exhaust fumes of motor vehicles

(Councell, et al., 2004; Malinowska, et al., 2015; McIntyre et al., 2014). Petroleum hydrocarbons

and heavy metals as cadmium, lead, zinc and copper are among the most common pollutants.

Some metals, as copper and zinc, are essential for living organisms but also potentially toxic at

elevated concentrations (Bossuyt and Janssen, 2005). Approximately 10% of the total particulate

Zn load in Sweden cities came from tire wear (Councell et al., 2004) and every year 10.000 tons

of rubber particles from tires end up in Swedish roads (Wik and Dave, 2005). High levels of zinc

or copper can cause reproductive, developmental, behavioral and toxic responses in diverse

aquatic organisms (Muyssen, et al., 2006). For example, in freshwater fish it has been observed

that the primary acute effect is an impaired branchial calcium influx that leads to hypocalcaemia

(Spry and Wood, 1985; and Muyssen and Jansson, 2002).

Toxic chemical contamination via runoff contributes to reduced species abundance, diversity

and increase the proliferation of non-native, pollution tolerant taxa (Spromberg et al., 2015).

Few studies have been performed on rain gardens and highway runoff, and the effect in different

species. Two species have been chosen to be studied: Daphnia magna or water fleas, a

planktonic crustacean, widely used in ecological and environmetal research and well known

(Ebert, 2005; Fryer, 1991). In addition, it is sensitive to low metals and dissolves toxines

concentrations (Guan and Wang, 2004; Muyssen et al., 2006).

As the second species a well known unicell green alga was used, Raphidocelis subcapitata

formerly known as Pseudokirchneriella subcapitata and Selenastrum capricornutum. It was

chosen among other reasons, for being widely used, easily available and easily maintained in the

laboratory under reproductible culture conditions. Microalgae play an important role in the

aquatic ecosystems and have shown being relatively sensitive to heavy metal or other toxicants

by decreasing growth rate and increasing cell size (Foster, 1977; Pica Granados et al., 2004).

1.1. Aim

The project aim was to study the effect of filtering water from several types of raingardens

watered with highway runoff and the runoff itself, using Daphnia magna and

Pseudokirchneriella subcapitata as bioindicators.

2. Method

2.1. Rain gardens

Eight rain gardens were constructed in March 2016 according to specifications in the

bioretention design manual from Prince George’s county, 2002. They were located in a terrace

at the top floor without any roof or side walls (Fig. 1). Polyethylene plastic boxes were used with

56x39x42cm size (65L). They were filled with 2 cm of gravel on the bottom and planting soil

according to European standards composition: light peat, dark peat, sand, lime, mineral

fertilizers, electrical conductivity (25%) 45 mS / m and pH (H2O) 5.5-6.5; until 4 cm to the top.

Holes were made at 2 cm from the bottom to collect the outflow into boxes underneath. Those

boxes underneath were also polyethylene plastic and sized 56x39x28 cm (45L). In order to avoid

any mixture between levels or any clogging of the holes, a mesh was settled between the soil

and the gravel (Fig. 3).

Figure 1. Rain gardens localization at a terrace at the top floor of a building without any roof or side walls, in order to simulate as much as possible natural condition.

Three species of perennial with high tolerance to dryness and salt stress were selected from a

study of system based approach to rain gardens (Wellander, 2015): Potentilla atrosanguinea

(commonly called Himalayan cinquefoil or ruby cinquefoil and native to mountains slopes at

lower elevation in the Himalayas), Armeria maritime ‘Alba’ (known as thrift or sea pink and

native from coastal areas across the Northern Hemisphere, especially in Europe and North Africa

to Turkey) and Sedum acre (commonly called common stonecrop or gold moss stonecrop and

native Europe).

In order to compare different types of rain garden composition, four treatments with two

replicates of each were performed (a sketch which clarifies the experimental design is shown

below in Fig 2 and 3):

1. Control: Soil + tapwater (ST), no vegetation, just soil and watered with tapwater.

2. Soil + runoff (SR): no vegetation and watered with highway runoff.

3. 6 Perennial (6P): 2 Potentilla atrosanguinea, 2 Armeria maritime ‘Alba’ and 2 Sedum acre

4. 3 Perennial (3P): 1 Potentilla atrosanguinea, 1 Armeria maritime ‘Alba’ and 1 Sedum acre

Figure 2 Rain gardens experimental design where the first pair of boxes are the controls which have just soil and they are watered with tapwater (ST). The second pair of boxes are composed of just soil and watered with highway run-off (SR). In the third pair of boxes two samples of Potentilla atrosanguinea, two samples of Armeria maritime ‘Alba’ and two samples of Sedum acre were planted and they were watered with highway runoff water (6P). In the last two boxes just one sample of each species were planted and it was watered with highway runoff water (3P). Armeria maritime ‘Alba’ is represented with a triangle, Sedum acre with a circle and Potentilla atrosanguinea with a square.

Figure 3 On the left side, a picture that shows the structure of the rain garden vertically: the two centimeters of gravel on the bottom, the mesh that avoids clogging of the holes that drain the water to the boxes underneath is shown with dot lines. On the right side, a picture of a six perennial rain garden from the top, which shows the distribution of two samples of Potentilla atrosanguinea, two samples of Armeria maritime ‘Alba’ and two samples of Sedum acre.

2.2. Highway runoff

During the months of March, April, May, June, August and September 2016, runoff water was

collected from two downspouts in both end sides of a bridge under the E6 highway in Mölndal,

Sweden (Fig. 4). The highway chosen is two lanes each way of 4,5m with 0,5m in end sides and

2 m between each way and it has a quite high traffic over the day with an average daily motor

vehicle traffic of 44510 vehicles per day in 2015 (http://vtf.trafikverket.se/SeTrafikfloden).

Figure 4. Location of the E6 highway in Mölndal (left) where the runoff was collected (GEO coordinates 57°38'49.2"N 12°01'21.6"E WGS84) and one of the downspouts in the bridge under the E6 highway (right)..

Taking into consideration the weather forecast, samples were collected after the first flush in

the different months. Daily and cumulative rainfall for each month of the study period is shown

in Appendix 1, with each storm water collection day (orange circle dots).

Highway runoff was collected in buckets placed at the outlet of the downspouts and promptly

transported to the laboratory where they were stored in a cold dark room in order to minimize

possible reactions.

The soil humidity condition of the rain garden affects how the filtered water infiltrate and will

also affect the dilution of the infiltrated water. It was important to maintain the condition of the

gardens as equal as possible during the measurement campaigns. Following method was

implemented. Rain gardens were considered moist when the 2 cm of the bottom of gardens had

at least 1 cm of water. In those days when they were drier, the gardens were watered with tap

water and it was waited until it had the same conditions. Watering testing days were chosen in

advance depending on the weather conditions, in order to avoid rain water. The amount of

runoff water was always between four and six liters and the filtered water was collected after

24 hours up to a maximum of 48 hours.

2.3. Toxicity tests

The possible toxicity of the highway runoff and the filtered water from the rain gardens were

analysed with two different procedures.

An acute toxicity immolibization assay with juvenile Daphnia magna extended to 72 hours was

performed according to the OECD guideline 202 (Organisation for Economic Co.operation and

Development, 2004). The purpose was to assess effects of chemicals towards mobility.

Guidelines suggested not to feed Daphnids as well as have pH and temperature under control

http://vtf.trafikverket.se/SeTrafikfloden

at the beginning of each test. These small freshwater crustaceans tolerate poor water quality,

pH between 6.5-9.5, temperature 18-22 °C. Highway runoff and the filtered rain garden water

from April, May, June, August and September were tested on Daphnia magna.

The experiment consists on using Petri dishes as test vessels and 10 young daphnids less than

24h old. They were exposed to 40mL of water solutions under room temperature conditions

(18-22°C). Immobilization (number of immobilized juveniles in each Petri dish) was recorded at

different time frames, 24, 48 and 72 hours and two replicates were done for each treatment.

A growth inhibition test with algae was carried out according to the OECD guideline 201.

Highway runoff water collected from the different months (April, May, June, August and

September 2016) were tested, in addition to the outflow from the raingardens from August and

September due to there was not enough filtered water from the other months to perform the

experiemnts. The purpose of this chronic test was to determine any effect on the growth of a

unicellular green algal species (Pseudokirchneriella subcapitata or Selenastrum capricornutum,

ATCC 22662). The initial cell concentration in the test cultures was around 104 cells/mL. Three

concentrations of runoff were arranged (100%, 50% and 25%) to be tested and four replicates

were included at each test concentration in all the months except June where there was just

water available for three replicates, in addition to four controls. Cell culture TC Easy Flasks,

sterile, 25 cm2 and 5-10ml working volume, were put in a GFL 3018 shaker at 50 rpm in the

climate room (21°±2°C). The cell concentration in each flask was determined at 72h after the

start of the test and in some cases also after a week. The algal medium used was MBL (Nichols,

H.W, 1973, for details see appendix 2) in miliQ water or runoff water according to the

concentration arranged to be tested, with pH 7´2. SkanIt Software 2.4.3. RE for Varioskan Flash

was used to measure the fluorometry (for protocol details appendix 3). In order to convert

relative fluorescence units (RFU) into number of cells per millimeter, a calibration curve was

used.

2.4. Chemical analyses

In order to find possible causes for the observed toxic effects, chemical analyses were planned.

However, only in August and September there was enough water left for implementing chemical

analyses. These were performed with the highway runoff and filtered water from August and

September by Eurofins Environment Testing Sweden AB. Total nutrients (P and N), total

suspended solids (TSS), total organic carbon (TOC), and heavy metals as manganese (Mn),

aluminum (Al), lead (Pb), cobalt (Co), copper (Cu), chromium (Cr), nickel (Ni) and zinc (Zn) were

analyzed.

2.5. Statistical analyses

Data were expressed as mean ± standard error. Mobility inhibition of Daphnia magna and

growth inhibition of Pseudokirchneriella subcapitata were analyzed using a one way analysis of

variance (ANOVA), followed by posthoc Bonferroni’s multiple range test. Statements of

significant differences were based on values p<0.05. All statistical analyses were conducted

using Statistical Packages for the Social Sciences (SPSS) Version 17.0.

3. Results and Discussion

3.1. Daphnia magna

In the experiments with Daphnia magna, testing any possible immobility caused by the water

filtered through rain gardens, no statistical significant differences were found in any of the

months. Some immobilized individuals were observed in filtered water but it could be attributed

to stress caused by handling the organisms. Therefore, D. magna does not seem to be affected

by the filtered water since no remarkable immobility was observed in any of the infiltrate from

the rain gardens.

However, on the highway runoff, statistical significance differences were found in August after

24h (p<0.001, F=84.868; appendix 4) and 48h (p<0.001, F=63.383; appendix 4), where the

mobility of the daphnids was significantly decreased by 55% and 70% respectively (Fig. 6). In

September mobility inhibition was observed after 48h with a 20% mobility reduced, but just

statistical significance differences were found after 72h (p<0.001, F= 18.758; appendix 4) with a

70% mobility reduced (Fig.7). The experiment was extended to 72h in all months due to during

the first months no alteration of the movement were observed, but some size differences were

observed. A dark dorsal line and bigger size was observed in most of the samples from filtered

water of raingardens (Fig.5). In those samples, the organisms seemed to move more actively and

faster than in those from runoff water which were smaller and moved seldom and in circles.

Those size differences were found mainly in the samples from May, June and August. In order

to know the cause of these size differences, a chronic toxicity test is recommended where the

food is controlled and starvation is discarded as possible cause.

Figure 5. Differences in size found in Daphnia magna after three days of exposure to filtered water through rain gardens and highway runoff from Mölndal (Sweden) in June 2016. On the left side, a sample on highway runoff water where it can be seen a light individual that showed smaller size and seldom movements. On the right side, a sample on filtered water where it can be seen a dark dorsal line and bigger that moved faster and more actively.

Figure 6. Acute immobilization test in Daphnia magna with highway runoff water from August 2016, Mölndal, Sweden after 24, 48 and 72h. Error bars represent the standard deviation.

Figure 7. Acute immobilization test in Daphnia magna with highway runoff water from September 2016, Mölndal, Sweden, after 24, 48 and 72h. Error bars represent the standard deviation.

3.2. Pseudokirchneriella subcapitata

Pseudokirchneriella subcapitata analysis were based on the inhibition of growth. Growth Inhibition was calculated as:

𝐺𝑟𝑜𝑤𝑡ℎ 𝐼𝑛ℎ𝑖𝑏𝑖𝑡𝑜𝑛 % = ( 1 − (𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑒𝑙𝑙𝑠 𝑖𝑛 𝑋 𝑠𝑎𝑚𝑝𝑙𝑒

𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑒𝑙𝑙𝑠 𝑖𝑛 𝑡ℎ𝑒 𝑐𝑜𝑛𝑡𝑟𝑜𝑙𝑠)) ∗ 100

Growth inhibition was observed in May, June and September, where statistical significance

differences were found at different concentrations and exposure time.

Figure 8 shows the inhibition in May after six days of exposure. The statistical analysis showed

that there was a significance inhibition at the highest concentration (Bonferroni test, p=0.000,

Std. Error = 0.340; appendix 4) with an average growth inhibition of 20%. A possible explanation

could be that, after three days of exposure, possible active regulation and storage mechanisms

were able to cope with certain pollutant concentration, but after six days of exposure at the

highest concentration, the limit of tolerance was reached and toxic effect were observed. In

addition, with the fact that in stressful situations, organisms invest more energy in survive and

consequently the biomass and growth rate is reduced.

0

20

40

60

80

100

24h 48h 72h

Mo

bili

ty In

hib

itio

n (

%)

Mobility inhibition test in Daphnia magna with highway runoff water from August

-40

-20

0

20

40

60

80

100

24h 48h 72h

Mo

bili

ty In

hib

itio

n (

%)

Mobility inhibition in Daphnia magna in contact with highway runoff water from September

Figure 9 shows the June inhibition data. Significant growth inhibition was obtained after three

days at 25% concentration (Bonferroni test, p=0.010, Std. Error= 0.044, appendix 4) where the

growth was inhibited in a 17´6% average. After four days exposure growth was inhibited in a

30% average at all concentrations and statistical significance differences were found, at 25%

(Bonferroni test, p=0.001, Std. Error= 0.058, appendix 4), 50% (Bonferroni test, p=0.004, Std.

Error= 0.058, appendix 4) and 100% concentration (Bonferroni test, p=0.002, Std. Error= 0.058,

appendix 4). This could be explained because low concentrations and three days of exposure,

were not enough for algae to implement all the mechanisms to cope with pollutants in the

water. It is known that some species are able to handle a range of certain concentration of

pollutants. In the presence of essential metals, like copper or zinc, each organism has an optimal

range of concentrations which depends on the homeostatic capacity and the natural

bioavailability background concentration (Bossuyt and Janssen, 2003). This

acclimation/adaptation capacity, that allows to regulate internal concentrations of metals, is

related to a general response (biological supply and demand) and to specific mechanisms (e.g.

metallothionein detoxification of metals). According to Bossuyt and Janssen (2005), the optimal

concentration range for copper for P.subcapitata is 1-35 µg/L and between 300-600 µg/L for

zinc, where there is an active regulation and storage mechanisms to regulate cellular levels.

Lower levels of those essential elements can lead to deficiency and higher levels to toxicity

(Berry and Wallace, 1981).

-20

-10

0

10

20

30

40

-10 10 30 50 70 90 110

Gro

wth

Inh

ibit

ion

(%

)

Runoff Concentration (%)

Growth Inhibition of Pseudokirchneriella subcapitata against highway runoff at 25, 50 and 100% concentration after six days of exposure, May 2016

Control

25

50

100

Figure 8. Growth Inhibition of Pseudokirchneriella subcapitata in highway runoff at 25, 50 and 100% concentration after six days of exposure, May 2016. Controls are referred as zero concentrations and each treatment have four replicates. Error bars represent the standard deviation.

Statistical significance differences was also found in September at the highest concentration

after three days of exposure (Bonferroni test, p=0.043, Std. Error= 0.146, appendix 4) and after

six days of exposure (Bonferroni test, p<0.001, Std. Error= 0.063, appendix 4). The growth was

reduced in a 40´4% average after three days of exposure and in a 63´8% after six days. Figure 10

shows the growth inhibition in September after three and six days of exposure. Therefore, it

seems that there is something in the water that at low concentrations does not affects algae

growth until a higher concentration is reached and the growth is significantly affected. This is a

pattern that basically every organism follows in contact with pollutants. In case of being

essential elements, there are established three phases: deficiency, tolerance and toxicity; and

in nonessential elements there is no deficient phase present (Berry and Wallace, 1981). In the

Figure 10, it can be observed a little negative growth at the lowest concentration even although

no statistical significance differences were found with the controls. This positive effect observed

at low concentration and the toxicity at the highest concentration may be related to the

presence of essential heavy metals in the highway runoff water as copper or zinc.

Figure 10 Growth Inhibition of Pseudokirchneriella subcapitata in highway runoff at 25, 50 and 100% concentrations after three and six days of exposure, September 2016. Controls are referred as zero concentrations and each treatment have four replicates. Error bars represent the standard deviation.

-10,0

0,0

10,0

20,0

30,0

40,0

50,0

-10 40 90

Gro

wth

Inh

ibit

on

(%

)


3 days of exposure, June 2016

-10,0

0,0

10,0

20,0

30,0

40,0

50,0

-10 40 90

Gro

wth

Inh

ibit

ion

(%

)


6 days of exposure, June 2016

Control

25%

50%

100%

Figure 9 Growth Inhibition of Pseudokirchneriella subcapitata in highway runoff at 25, 50 and 100% concentrations after three and six days of exposure, June 2016. Controls are referred as zero concentration and each treatment have three replicates. Error bars represent the standard deviation.

-80,00

-60,00

-40,00

-20,00

0,00

20,00

40,00

60,00

80,00

-10 40 90

Gro

wth

Inh

ibit

ion

(%

)


3 days of exposure, September 2016

-80,00

-60,00

-40,00

-20,00

0,00

20,00

40,00

60,00

80,00

-10 40 90

Gro

wth

Inh

ibit

ion

(%

)


6 days of exposure, September 2016

Controls

25%

50%

100%

In the filtered highway runoff water through the rain gardens from August and after three days

of exposure against Pseudokirchneriella subcapitata, statistical significance differences were

found (p<0.001, F= 9.056, appendix 4). Soil +runoff (SR) was highly significant different from the

control and the other types of rain gardens. It is observed a negative inhibition, which means an

increase of growth comparing with any of the treatments (Fig.11). In the other treatments there

is barely no inhibition of growth.

Figure 11. Growth Inhibition percentage of Pseudokirchneriella subcapitata against rain gardens outflow, watered with highway runoff after three days of exposure and highway runoff water, August 2016. There are two replicates of each treatment and three samples were measured from each. Error bars represent the standard deviation.

After six days of exposure, inhibition can be seen in all the rain gardens and statistical

significance differences were found in 6P, six perennial (3x2) plants watered with pure runoff

(Fig. 12). A possible explanation is that it could be a toxicity effect where it has reached the limit

of tolerance of an element or elements is reached and all the regulation or storage mechanisms

could be failing.

Figure 12. Growth Inhibition percentage of Pseudokirchneriella subcapitata against rain gardens outflow watered with highway runoff after six days and highway runoff water, August 2016. There are two replicates of each treatment and three samples were measured from each. Error bars represent the standard deviation.

-120,0

-100,0

-80,0

-60,0

-40,0

-20,0

0,0

20,0

40,0

1 2 3 4 5 6 7 8

Control Soil+Tapwater Soil + Runoff Perennial (3x2) +runoff

Perennial (3x1) +runoff

Runoff

Gro

wth

Inh

ibit

ion

%

Treatment

-10,0

0,0

10,0

20,0

30,0

40,0

50,0

1 2 3 4 5 6 7 8



Runoff

Gro

wth

Inh

ibit

ion

%

Treatment

In September the same tendency is followed after three and six days of exposure. On filtered

soil+runoff rain gardens appeared a considerable growth after three days of exposure (Fig.13)

and statistical significance differences were found with the other treatments (Bonferroni test,

p<0.001, Std. Error= 0.305, appendix 4). No inhibition or growth was observed in any of the other

treatments after three days of exposure as well. After six days of exposure, Figure 13, there was

growth inhibition and statistical significance in soil+runoff (Bonferroni test, p=0.041, Std. Error=

0.063, appendix 4) and six perennials (Bonferroni test, p=0.038, Std. Error= 0.063, appendix 4) .

Figure 13. Growth Inhibition percentage of Pseudokirchneriella subcapitata against rain gardens outflow watered with highway runoff after three days of exposure and highway runoff water, September 2016. There are two replicates of each treatment and three samples were measured from each. Error bars represent the standard deviation.

Figure 14. Growth Inhibition percentage of Pseudokirchneriella subcapitata against rain gardens outflow watered with highway runoff after six days and highway runoff water, September 2016. There are two replicates of each treatment and three samples were measured from each. Error bars represent the standard deviation.

When looking at the remediation effect of different types of rain gardens, no differences were

found concerning the number of plants. All rain gardens worked as good biofilter for runoff

water, although the raingarden with just soil and watered with highway runoff seemed to

provide more benefits to algae. That could be due to, even although there was a filtration of

pollutants observed, higher amount of compounds are not retained by the soil and go through

all the layers. In the short term, P. subcapitata is benefited by those substances or elements. But

-500,0

-400,0

-300,0

-200,0

-100,0

0,0

100,0

200,0

1 2 3 4 5 6 7 8



Runoff

Gro

wth

Inh

ibit

ion

%

Treatment

-20,0

-10,0

0,0

10,0

20,0

30,0

40,0

50,0

60,0

70,0

80,0

90,0

1 2 3 4 5 6 7 8



Runoff

Gro

wth

Inh

ibit

ion

%

Treatment

after longer period of exposure, as previously mentioned, all the regulation mechanisms to cope

with pollutants seemed to be saturated and a significant inhibiton of P. subcapitata growth rate

occurred.

In the experiments with P. subcapitata the rain gardens watered with tap water was included.

P. subcapitata also showed inhibition after six days of exposure of water from the gardens in

both August and September. The reason could be that something in the rain gardens could affect

these organisms apart from the runoff itself. But, due to, the infiltration rate was not measured

and the water was collected in the boxes underneath after several hours or the next day, it could

also be accumulation of pollutants in the layers between experiments and it could explain the

high concentration of the heavy metals in August comparing with the runoff itself.

3.3. Chemical Analyses

The chemical analyses show that in August, the runoff concentration in almost all the

compounds analyzed is lower or in the same range of concentration than the filtered water

concentrations (Table 1 and 2 and Figure 15). Just on Zn, the levels are significantly higher in the

runoff than in the rain gardens. Even then, no effect was observed in any of the filtered water

samples from August but in the runoff on Daphnia magna. In all the filtered water from

September there is a decrease in the pollutants concentration comparing with the runoff

concentration (Table 1, 2 and figure 15). The heavy metals data were compared with water

quality criteria for lakes and watercourse from Swedish EPA (Environmental Protection Agency)

and categorized in any of the five classes (Table 2). According to these values, high concentration

of copper and zinc was found in the highway runoff water tested in August and very high levels

of copper and zinc in September. Those levels, high and very high, are considered as a growing

risk of biological effects where the survival of aquatic organisms is affected even in short- term

(EPA, Water Standars criteria). In addition, moderate high concentration of chromium and lead

are found in September, which means that effects may occur. Also, according to Bossuyt and

Janssen (2005), adverse effects were observed at copper concentrations of 0.1 mg/L in

Pseudokirchneriella subcapitata and 0.15 mg/L in Daphnia magna. These values agree with the

results in the experiment, where a maximum of 0.12 mg/L of copper was found in the highway

runoff water from September and statistical significance differences were found for

Pseudokricheriella Subcapitata but not for Daphnia magna. Therefore, one of the compounds

that could be affecting the growth of Pseudokrichneriella subcapitata in September could be

copper.

In addition, the zinc concentration is 1 mg/L in September, above the limit of tolerance for the

organisms according to EPA and Muyssen and Janssen (2002), whose stablished an optimal

range concentration of 300-600µg/L for Daphnia magna.

Several authors report an effect of acclimatation on the acute and chronic toxicity of zinc, copper

and cadmium to Daphnia magna and Pseudokirchneriella subcapitata (Foster, 1977; Griffiths,

1980; Le Blanc, 1982; Boder et al. 1990; Stuhlbacher et al., 1992, Muyssen and Janssen, 2001;

Bossyt and Janssen, 2003 and 2004; Guan and Wang, 2014). These organisms can regulate some

essensial metals that are potentially toxic at high concentrations in order to maintain them in a

optimal concentration range. Each species have diverse mechanisms to compete with stress.

The concetrations which species can acclimate or adapte to, depend on the background

concentrations in their ecosystem and the specific mechanisms that allows to active regulation

or storage(Bossyt and Janssen, 2003 and 2004).

http://www.swedishepa.se/

Table 1. Chemical results of the highway runoff water and filtered through raingardens from August and September 2016. There are four types of rain gardens: those with soil and watered with tapwater (controls); those with just soil but watered with runoff; those with three perennial species planted and watered with runoff; and those with two samples of three perennial species and watered with runoff. TSS, total suspended solids; TOC, total organic carbon; P, Phosphorus; N, Nitrogen; Mn, manganese; Al, Aluminum; Pb, Lead; Co, Cobalt; Cu, Copper; Cr, Chromium; Ni, Nickel and Zn, Zinc.

August September

Treatment Control Soil 3 perennials 6 perennials Runoff Control Soil 3 perennials 6 perennials Runoff

Ch

em

ical

an

alys

is (

mg/

l)

TSS - - - - 6,1 13 28 18 23 52

TOC 42 50 110 80 6 41 61 96 87 22

P 6 4,3 11 8,5 0,031 6 4,3 13 8,7 0,1

N 1,6 3,6 6,6 3,9 2 2 7,9 8 13 8,5

Mn 0,0084 0,019 0,034 0,038 0,0033 0,0089 0,027 0,023 0,028 0,15

Al 0,21 0,53 1 1,1 0,029 0,18 0,68 0,61 0,86 0,6

Pb 0,0011 0,0012 0,0017 0,002 <0,0005 <0,0005 0,002 0,0014 0,0022 0,0019

Co <0,001 <0,0010 0,0014 0,0014 <0,001 <0,001 0,001 <0,0010 0,0011 0,0044

Cu 0,036 0,028 0,031 0,03 0,025 0,015 0,039 0,054 0,045 0,12

Cr <0,001 0,0028 0,0059 0,007 0,0015 <0,001 0,0055 0,0046 0,0051 0,0077

Ni 0,0073 0,0072 0,02 0,017 0,0012 0,0049 0,0064 0,01 0,012 0,0061

Zn 0,015 0,061 0,1 0,066 0,23 0,015 0,081 0,06 0,1 1

Table 2. Assessment of metals in water according to Swedish EPA (appendix 5).

Class Description Colour

1 Very low conc.

2 Low conc.

3 Mod. High. Conc.

4 High conc.

5 Very high conc

All the results could not be explained based on the chemical analyses, being important to take

into consideration other pollutants very common in the highway runoff, like polyaromatic

hydrocarbons (PAHs) (Wik and Dave, 2005). PAHs are compounds used in the manufacturing of

car tires and very toxic for aquatic organisms. Approximately 10.000 tonnes of rubber particles

from tires end up in swedish roads and each tires has around one kilograme of high aromatic

oils (Wik and Dave, 2005).

August and September were the driest months under the study, with 12 and 20 days without

rain respectively (appendix 1). In these two months greater differences are found in both species

under this study, and that could be justified due the accumulation of pollutants during this long

periods without rain. Moreover, according to Wik and Dave (2005), summer tires are more toxic

than winter tires for having higher amount of PAHs.

Chemical Analysis

0

10

20

30

40

50

60

Control Soil 3perennial

6perennial

Runoff

mg/

l

Total Suspended Solids

0

2

4

6

8

10

12

14


6perennial

Runoff

mg/

l

Phosphorus

0

2

4

6

8

10

12

14

Control Soil 3Perennial

6Perennial

Runoff

mg/

l

Nitrogen

00,020,040,060,08

0,10,120,140,16


6perennial

Runoff

mg/

l

Manganese

0

0,2

0,4

0,6

0,8

1

1,2


6perennial

Runoff

mg/

l

Aluminium

0

20

40

60

80

100

120


6perennial

Runoff

mg/

l

Total Organic Carbon

Figure 15. Concentration of TSS, TOC, Phosphorous, Nitrogen and heavy metals in highway runoff and filtered raingardens water from August and September, 2016 in Mölndal, Sweden. Type of raingardens under study: soil watered with tapwater as controls; soil watered with runoff; one individual of 3 types of perennials planted and watered with runoff: Potentilla atrosanguinea, Armeria maritime ‘Alba’ and Sedum acre; two individuals of each of the same species previously mentioned and watered with runoff and the runoff itself. Blue line represents August and orange line represents September.

0

0,0005

0,001

0,0015

0,002

0,0025


6perennial

Runoff

mg/

lLead

0

0,001

0,002

0,003

0,004

0,005


6perennial

Runoff

mg/

l

Cobalt

0

0,002

0,004

0,006

0,008

0,01


6perennial

Runoff

mg/

l

Chromium

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14


6perennial

Runoff

mg/

l

Copper

0

0,005

0,01

0,015

0,02

0,025


6perennial

Runoff

mg/

l

Nickel

0

0,2

0,4

0,6

0,8

1

1,2


6perennial

Runoff

mg/

l

Zinc

4. Conclusions

Highway storm water is chemically complex and contains unidentified compounds that could be

removed using inexpensive bioinfiltration as rain gardens. In this study rain gardens have been

proven to be efficient biofilters that prevent lethal impacts on Daphnia magna and

Pseudokirchneriella subcapitata. Even though pollutants were not completely removed by

infiltration just using soil, an improvement in the quality of the water was observed. But it is also

important to mention that the removal efficiency varies for the different compounds.

In this study, the mobility of Daphnia magna is affected by the runoff itself in August and

September on the acute toxicity, but not for the filtered water through the rain gardens and

those with just soil. August and September were the driest months under study, and the runoff

collected were the flush after longer periods without rain.

Some size differences were also observed in Daphnia magna in May, June and August, but a

chronic toxicity test is needed to clarify the cause and not mix it up with starvation.

An inhibition effect on the growth of Pseudokirchneriella subcapitata was observed in May, June

and September but the effect varies depending on the concentration and time of exposure. The

only filtered water tested in Pseudokirchneriella subcapitata were from August and September.

There was a tendency where a positive growth was observed in a short run and an inhibition of

growth appeared after six days of exposure. The inhibition could be due to the toxicity of the

runoff itself, but also to something in the material used in the construction of the rain gardens

or even a possible accumualtion in the layers between the experiments.

High levels of zinc and copper were found in the highway runoff from August and

September,even moderate high levels of chromium and lead were found in September too.

These metals are potentially toxic for the organisms at high concentrations, but through rain

gardens they are reduced and lethal impact on Daphnia magna and Pseudokirchneriella

subcapitata were prevented at least in a short run.

To get more accurate results of the drain capacity it would be recommended to scale the

experiments and locate the raingardens in the sides of the road or other urban areas with heavy

traffic, for example parking lots and bus stations. It would be required to measure the infiltration

rate in order to avoid any possible accumulation in the rain gardens along the different sampling.

In future studies, it would be of interest to track highway runoff during longer periods of time,

to compare through different seasons and different years. At least during winter and the

snowmelt period where the runoff duration and event volume increases, and it is expected a

higher concentration of pollutants.

Due to car tires contain several water soluble compounds that have severe effects to aquatic

organisms, in particular, polyaromatic hydrocarbons (PAHs), it is important to take it into

consideration in the chemical analyses in further studies.

5. References

Berry, W. L. and Wallace, A. Toxicity: The concept and relationship to the dose response curve.

Journal of Plant Nutrition. 3: 1-4, 13-19.

Bossuyt, B. T. A. and Jansses, C. R. 2003. Acclimation of Daphnia magna to environmentally

realistic copper concentrations. Comparative Biochemistry and Physiology Part C 136: 253-264.

Bossuyt, B. T. A. and Jansses, C. R. 2004. Long-term acclimation of Pseudokirchneriella

subcapitata (Korshikov) Hindak to different copper concentrations: changes in tolerance and

physiology. Aquatic Toxicology. 68: 61-74.

Bossuyt, B. T. A. and Janssen, C. R. 2005. Copper regulation and homeostasis of Daphnia magna

and Pseudokirchneriella subcapitata: influence of acclimation. Environmental Pollution 136:

135-144.

Bray, B., Gedge, D., Grant, G. and Leuthvilay, L. Rain Garden Guide, UK. 2012. Environment

Agency.

Councell, T. B., Duckenfield, K. U., Landa, E. R. and Callender, E. 2004. Tire-Wear particles as a

source of Zinc to the environment. Environ. Sci. Technol. 38: 4206-4214.

Dietz, M.E. and Clausen, J.C. 2005. A field evaluation of rain garden flow and pollutant treatment.

Water, Air, and Soil Pollution. 167:123-138.

Ebert, D. 2005. Ecology, Epidemiology, and Evolution of Parasitism in Daphnia

[Internet].Bethesda (MD): National Library of Medicine (US), National Center for Biotechnology

Information. Available from: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Books

Swedish Environmental Protection Agency. Environmental Quality Criteria. Lakes and Watercourses. Report 5050. ISBN 91-620-5050-8. ISSN 0282-7298. Foster, P.L. 1977. Copper exclusion as a mechanism of heavy metal tolerance in a green alga.

Nature. 269: 322-323.

Fryer, G. 1991. Functional morphology and the adaptive radiation of the Daphniidae

(Branchiopoda: Anomopoda) Philes Trans R. Soc 331: 1-99.

Griffiths, P. R. E. 1980. Morphological and Ultrastructural Effects of Sublethal Cadmium

Poisoning on Daphnia. Environmental Research. 22:277-284.

Guan, R and Wang, W-X. 2004. Cd and Zn Uptake Kinetics in Daphnia magna in Relation to Cd

Exposure History. Environ. Sci. Technol. 38: 6051-6058.

Heijerick, D. G., Janssen, C. R. and De Coen, W. M. 2002. The Combined Effects of Hardness, pH

and Dissolved Organic Carbon on the Chronic Toxicity of Zn to D. magna: Development of a

Surface Response Model. Arch. Environ. Contam. Toxicol. 44: 210-217.

Hinman, C. 2013. Rain Garden Handbook for Western Washington. A Guide for Design,

Maintenance and Installation. Department of Ecology State of Washington. Washington State

University Extension.

http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Books

Katsumata, M., Koike, T., Nichikawa, M., Kazumura, K. and Tsuchiya, H. 2006. Rapid

ecotoxicological bioassay using delayed fluorescence in the green alga Pseudokirchneriella

subcapitata. Water research 40 (18): 3393-4000.

Malinowska, E., Jankowski, K., Wisniewska-Kadzajan, B., Sosnowski, J., Kolczarek, R., Jankowska,

J. and Ciepiela, G.A. 2015. Content of Zinc and Copper in Selected Plants Growing Along a

Motorway. Bull Environ Contam Toxicol. 95: 638-643.

McIntyre, J. K., Davis, J.W., Incardona, J.P., Stark, J.D., Anulacion, B. F. and Scholz, N.L. Zebrafish

and clean water technology: Assessing soil bioretention as a protective treatment for toxic urban

runoff. Science of the Total Environment. 500-501: 173-180.

Moreira-Santos, M., Soares, A. M. V. M. and Ribeiro, R. 2004. An in situ bioassay for freshwater

environments with the microalga Pseudokirchneriella subcapitata. Ecotoxicology and

Environmental Safety. 59: 164-173.

Muyssen and Janssen. 2002. Accumulation and regulation of Zinc in Daphnia magna: Links with

Homeostasis and Toxicity. Arch. Environ. Contam. Toxicol. 43: 492-496.

Muyssen, B. T. A., De Schamphelaere, K. A.C. and Janssen, C. R. 2006. Mechanisms of chronic

waterborne Zn toxicity in Daphnia magna. Aquatic Toxicology 77: 393-401.

Nichols, H. W. 1973. Handbook of Phycological Methods- Ed. J. R. Stein. 16-17 pp. Cambridge

University Press.

OECD Guidelines for testing of Chemicals. Guideline 202: Daphnia sp., Acute Immobilization

Test, adopted April 2004.

OECD Guidelines for testing of Chemicals. Guideline 201: Alga, Growth Inhibition Test, adopted

June 1984.

Pica Granados, Y., Ronco, A. and Díaz Báez, M. C. 2004. Ensayo de toxicidad crónica con el alga

Selenastrum Capricornutum (Pseudokirchneriella subcapitata) por el método de enumeración

celular basado en el uso de hemocitómetro Neubauer.

Spromberg, J.A., Baldwin, D.H., Damm, S. E., McIntyre, J.K., Huff, M., Sloan, C.A., Anulacion, B.F.,

Davis, J.W. and Scholz, N.L. Coho salmon spawner mortality in western US urban watersheds-.

Bioinfiltration prevents lethal storm water impacts. Journal of Applied Ecology. 12534: 1365-

2664.

Valtanen, M., Sillanpää, N. and Setälä, H. 2015. Key factors affecting urban runoff pollution

under cold climatic conditions. Journal of Hydrology. 529: 1578-1589.

Wellander, Å. 2015. Systembeskrivning av regnbäddar. Från ståndortsuppbyggnad till

växtfysiologiska och morfologiska egenskaper. Master project in Landscape Architecture. SLU,

Sverige lantbruksuniversitet.

Wik, A. and Dave, G. 2005. Environmental labeling of car tires - toxicity to Daphnia magna can

be used as a screening method. Chemosphere 58: 645-651.

6. Appendices

6.1. Appendix 1

Precipitation histograms for April, May, June, August and September from the closest station to the runoff water collecting place. Kållered D, number 72360

(http://www.smhi.se/klimatdata/meteorologi/nederbord). Orange dots are the collecting day in each month.

4,5

0

5

10

15

20

25

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29

Dai

ly r

ain

fall

(mm

)

Day of the month

Mean day rain precipitation in April 2016 in Kållered (mm)

0,5

0

0,5

1

1,5

2

2,5

3

3,5

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31D

aily

rai

nfa

ll (m

m)

Day of the month

Mean day rain precipitation in May 2016 in Kållered (mm)

5,4

0

5

10

15

20

25

30

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31

Dai

ly r

ain

fall

(mm

)

Day of the month

Mean day rain precipitation in June 2016 in Kållered (mm)

25,7

0

5

10

15

20

25

30

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31

Dai

ly r

ain

fall

(mm

)

Day of the month

Mean day rain precipitation in August 2016 in Kållered (mm)

1,50

5

10

15

20

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29

Dai

ly r

ain

fall

(mm

)

Day of the month

Mean day rain precipitation in September 2016 in Kållered (mm)

http://www.smhi.se/klimatdata/meteorologi/nederbord

6.2. Appendix 2

MBL Medium

Reference: Nichols, H. W. (1973) in Handbook of Phycological Methods, Ed. J. R. Stein, pp. 16-

17. Cambridge University Press.

Online source:

http://www.marine.csiro.au/microalgae/methods/Media%20CMARC%20recipes.htm#MBL

Adapted by the CSIRO for freshwater algae

Stock solutions Per litre distilled water

1. CaCl2.2H2O 36.76 g

2. MgSO4.7H2O 36.97 g

3. NaHCO3 12.60 g

4. K2HPO4 8.71 g

5. NaNO3 85.01 g

6. Na2SiO3.9H2O 28.42 g

7. Na2EDTA 4.36 g

8. FeCl3.6H2O 3.15 g

9. Metal Mix Add each constituent separately to ~750mL

of distilled H2O, fully dissolving between

aditions. Finally make up to 1L with distilled

H2O.

CuSO4.5H2O 0.01 g

ZnSO4.7H2O 0.022 g

CoCl2.6H2O 0.01 g

MnCl2.4H2O 0.18 g

Na2MoO4.2H2O 0.006 g

10. Vitamin stock

Cyanocobalamin (Vitamin B12) 0.0005 g / L dH2O

Thiamine HCl (Vitamin B1) 0.10 g / L dH2O

Biotin 0.0005 g / L dH2O

11. Tris stock 250.0 g / L dH2O

Store all stock solutions in the refrigerator.

To Prepare MBL Medium Add 1mL of each stock solution (1 – 11) to 1litre distilled water. (For species which cannot use nitrate substitute 1mL of NH4Cl made up to 5.4 g /L H2O) Adjust pH to 7.2 with HCl. Autoclave at 121°C (15PSI for 15 mins). Downloaded from seniorphysics.com/biol/eei.html

http://www.marine.csiro.au/microalgae/methods/Media%20CMARC%20recipes.htm#MBL

6.3. Appendix 3

Start SkanIt RE for varioskan Flash 2.4.3.

Fluorometric:

Excitation wave (nm): 425

Emission wavelength (nm): 680

Excitation bandwidth (nm): 12

Dynamic range: Auto range

Measurement time (ms): 100

6.4. Appendix 4

Statistical Analysis

In the next tables the significant cases found in the analyses are shown:

1. Acute Immobilization Toxicity Test in Daphnia magna:

a. August

Univariate Analysis of Variance

Dependent Variable: Mobility

Exposure Time

Variable F Sig. R Squared

Adjusted R Squared

24h Treatment 83´868 0´000 0´963 0´951

48h Treatment 63´383 0´000 0´948 0´933

72h Treatment 139´262 0´000 0´979 0´972

Post Hoc Tests: Bonferroni Mutiple Comparissons 95% Confidence Interval

Exposure time

Treatment Treatment Mean Difference

Std. Error

Sig Lower Bound

Upper Bound

24h Runoff Soil+Tapwater (control) 0´55 0´034 0´000 0´44 0´66

Soil+Runoff 0´53 0´034 0´000 0´41 0´64

Perennial3 0´53 0´034 0´000 0´41 0´64

Perennial6 0´55 0´034 0´000 0´44 0´66


Soil+Runoff 0´65 0´049 0´000 0´49 0´81

Perennial3 0´68 0´049 0´000 0´51 0´84

Perennial6 0´70 0´049 0´000 0´54 0´86


72h Soil+Runoff 0´65 0´049 0´000 0´49 0´81

Perennial3 0´68 0´049 0´000 0´51 0´84

Perennial6 0´70 0´049 0´000 0´54 0´86

b. September


Dependent Variable: Mobility

Exposure Time

Variable F Sig. R Squared

Adjusted R Squared

72h Treatment 18´758 0´000 0´862 0´816

Post Hoc Tests: Bonferroni Mutiple Comparissons

95% Confidence Interval

Time exposure


Std. Error

Sig Lower Bound Upper Bound

72h Runoff Soil+Tapwater (control) 0´60 0´094 0´000 0´28 0´92 Soil+Runoff 0´70 0´089 0´000 0´39 1´01 Perennial3 0´70 0´089 0´000 0´39 1´01 Perennial6 0´63 0´089 0´000 0´32 0´93

2. Growth Inhibition Test in Pseudokirchneriella subcapitata:

a. April and May


Dependent Variable: Growth Inhibition

Exposure Time

Variable F Sig. R Squared Adjusted R Squared

6 days-Aprils Runoff Concentration 4´728 0´021 0´542 0´427 April

6 days-May

22´759 0´000 0´851 0´813 May


95% Confidence Interval Treatment Treatment Mean

Difference Std. Error

Sig Lower Bound

Upper Bound

Exposure Time

Runoff Concentration Runoff Concentration

April 0´5 1 0´13650 0´39527 0´029 0´26111 0´1188

May 1 0 0´21132 0´034012 0´000 0´10409 0´31855 0´25 0´16926 0´034012 0´002 0´06203 0´27649 0´5 0´26574 0´034012 0´000 0´15851 0´37297

b. June



Exposure Time


72h Runoff Concentration 5´667 0´022 0´680 0´560

96h

13´521 0´002 0´835 0´773

Post Hoc Tests:Bonferroni Mutiple Comparissons



Sig Lower Bound

Upper Bound

Exposure Time

Runoff Concentration

72h 0 0´25 0´17629 0´043835 0´023 0´32878 0´02379

96h 0 0´25 0´31911 0´057877 0´003 0´52046 0´11776 0´5 0´27522 0´057877 0´009 0´47657 0´07387 1 0´30208 0´057877 0´005 0´50343 0´10073

c. September



Exposure Time


72h Runoff Concentration 8´669 0´002 0´684 0´605

6 days

47´547 0´000 0´922 0´903


95% Confidence Interval


Std. Error

Sig Lower Bound

Upper Bound

Exposure Time

Runoff Concentration

72h 1 0´25 0´71566 0´146270 0´002 0´25452 1´17680 0´5 0´53935 0´146270 0´019 0´07820 1´00049

6 days 1 0 0´63750 0´62503 0´000 0´44045 0´83456 0´25 0´65005 0´62503 0´000 0´45300 0´84710 0´5 0´48386 0´62503 0´000 0´28680 0´68091

d. Raingardens- August



Exposure Time


72h Treament 9´056 0´000 0´622 0´553

6 days

4´196 0´011 0´433 0´330




Sig Lower Bound

Upper Bound

Exposure Time

72h Soil+Runoff Control 0´53935 0´133733 0´006 0´95644 0´12226 Soil+Tapwater 0´54809 0´109193 0´001 0´88864 0´20754

Perennial 3 0´51729 0´109193 0´001 0´85784 0´17673 Perennial 6 0´50440 0´109193 0´001 0´84496 0´16385

6 days Control Soil+Tapwater 0´26016 0´070000 0´012 0´47848 0´04184

e. Raingardens- September



Exposure Time


72h Treament 22´770 0´000 0´805 0´770

1 week

4´680 0´007 0´460 0´361




Sig Lower Bound

Upper Bound

Exposure Time

72h Soil+Runoff Control 2´53220 0´373737 0´000 3´69782 1´36658 Soil+Tapwater 2´25013 0´305155 0´000 3´21185 1´30840 Perennial 3 2´45268 0´305155 0´000 3´40441 1´50096 Perennial 6 2´10827 0´305155 0´000 3´05999 1´15654

1 week Control Soil+Runoff 0´21023 0´062581 0´028 0´40541 0´01505 Soil+Tapwater 0´20062 0´062581 0´041 0´39580 0´00544 Perennial 6 0´20260 0´062581 0´038 0´39778 0´00742

6.5. Appendix 5

Water quality Criteria for lakes and watercourses according to Swedish Environmental

Protection Agency

Assessment of current conditions:

Assessment of deviation from references values

the use of rain gardens as biofilters in highways and … · rain gardens, also called bioretention...

Documents