Journal of Environmental Management 152 (2015) 171e176
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
A pilot study to evaluate runoff quantity from green roofs Ju Young Lee a, Min Jung Lee c, Mooyoung Han b, * a
KIST(Korea Institute of Science and Technology)-Natural Products Research Center, Gangnueng, 210-340, South Korea Department of Civil and Environmental Engineering, Seoul National University, 599 Gwanak-ro, Gwanak-gu, Seoul, 151-744, South Korea c LIFTRC(LED-IT Fusion Technology Research Center), Daegu, South Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 January 2013 Received in revised form 24 July 2014 Accepted 17 January 2015 Available online
The use of green roofs is gaining increased recognition in many countries as a solution that can be used to improve environmental quality and reduce runoff quantity. To achieve these goals, pilot-scale green roof assemblies have been constructed and operated in an urban setting. From a stormwater management perspective, green roofs are 42.8e60.8% effective in reducing runoff for 200 mm soil depth and 13.8 e34.4% effective in reducing runoff for 150 mm soil depth. By using Spearman rank correlation analysis, high rainfall intensity was shown to have a negative relationship with delayed occurrence time, demonstrating that the soil media in green roofs do not efficiently retain rainwater. Increasing the number of antecedent dry days can help to improve water retention capacity and delay occurrence time. From the viewpoint of runoff water quality, green roofs are regarded as the best management practice by filtration and adsorption through growth media (soil). © 2015 Elsevier Ltd. All rights reserved.
Keywords: Green roofs Environmental quality Reducing runoff quantity Spearman rank correlation Best management practice (BMP)
1. Introduction Recently, green roofs have been receiving increased recognition in many countries such as the USA, Japan, and European countries (Vijayaraghavan et al., 2012; Berndtsson et al., 2008; Mentens et al., 2006). Particularly, the installation of green roofs is now increasing in Korea, as shown Fig. 1. Green roofs have several positive effects in the urban setting. The most important effect is their ability to retain and detain rainwater (Villarreal and Bengtsson, 2005). Green roofs have been introduced for reducing urban stormwater runoff from rooftops. The growth media in the green roof system typically retains and detains rainwater. The depth of the growth media can play an important role when used in an eco-friendly drainage system by slowing and reducing runoff volume. Retaining rainwater in soil reduces peak flow, which means that it prolongs the time of concentration (Bengtsson et al., 2005). The green roof has several benefits that include reduction of the urban heat island effect (Wong et al., 2003) and reducing building energy consumption by cooling roofs during the summer (Vijayaraghavan et al., 2012; Del Barrio, 1998). Finally, green roofs improve urban biodiversity and create habitat for plants and animals in addition to their
* Corresponding author. E-mail addresses: [email protected] (J.Y. Lee), [email protected] (M.J. Lee), [email protected] (M. Han). http://dx.doi.org/10.1016/j.jenvman.2015.01.028 0301-4797/© 2015 Elsevier Ltd. All rights reserved.
esthetic appeal (Vijayaraghavan et al., 2012; Emilsson et al., 2007). Green roofs are mainly divided into two types. Intensive green roofs are characterized by a thick layer of growth medium of more than 200 mm thickness, and extensive green roofs have a layer of less than 200 mm thickness (Nagase and Dunnett, 2010). Most green roof systems consist of a drainage layer, a root barrier, and a waterproof membrane. Previous studies have focused on evaporative and radiative heat transfer mechanisms of the different types of roof assemblies (Ayata et al., 2011; Tabares-Velasco and Srebric, 2009). Many previous studies have considered the runoff quality from green roofs (Ayata et al., 2011; Berndtsson et al., 2008, 2006). These studies show that green roofs can filter and absorb nonpoint pollutants (Berndtsson et al., 2008, 2006). However, in the longterm they can contribute to the deterioration of runoff water quality by releasing fertilizers regarded as pollutants (Berndtsson, 2010; Teemusk and Mander, 2007; Berndtsson et al., 2006). Moran et al. (2003) found that green roof systems have advantages and disadvantages. They are the best management practice for achieving water retention and peak flow reduction as benefits; however, they result in high nutrient concentrations. Thus, the main purpose of the present paper is to evaluate the reduction of peak flow, which prolongs the time of concentration during rainfall events.
172
J.Y. Lee et al. / Journal of Environmental Management 152 (2015) 171e176
Fig. 1. Extensive green roof (NAM-SAN annex, Seoul City Hall, Korea). Table 1 Outline of each test batch. Material
A B C D
Size (cm)
Acryl Concrete Green roof
Width
Length
Height
70 100 100
70 100 100
e 3 15 20
Soil
Vegetation
Purpose
Measure list
Note
e e Perlite
e e Sedum
Rainfall sampling Runoff sampling Green roof outflow
W$Qual, W$Quan W$Qual, W$Quan
Fig. 1(a) Fig. 1(b) Fig. 1(c)
W$Qual: Water Quality; W$Quan: Water Quantity.
2. Methods and materials 2.1. Data collection The quantity and quality of runoff were evaluated for 4 types of pilot facilities that used acryl, concrete, and 2 different green roof models, and were established on the rooftop of Seoul National University Building 35, Gwanak-Gu, Seoul, Korea (see Table 1). There were 7 rainfall events in the test period from May 7, 2011 to Sept 29, 2011. Total rainfall data and rainfall intensities were obtained for Seoul from the Seoul Meteorological Office (see Table 2). A Norton Rainfall Simulator with 2 heads, model (DIK-600), which was developed by USDA-ARS NSERL at Purdue University, West Lafayette, Indiana, was used to simulate the 7 rainfall events. For harvesting rainwater unaffected by the catchment area, the catchment facility was made of acrylic material (Fig. 2(a)). The catchment installed in test batch B was made of concrete to examine the effect of the catchment on runoff quality (Fig. 2(b)). In Fig. 2(c), batches C (150 mm) and D (200 mm) were sampled at different soil depths to examine the relationship of runoff quantity
and water retention with soil depth. Further, Fig. 2(c) shows a cross section of the test batches C and D. These test batches were composed of layers of sedum, volcanic materials and soil with peat moss (50 mm), perlite (100 mm for test C and 150 mm for test D), and a drainage plate (40e50 mm). The soils were provided by the producer (GreenInfra, Co., Ltd., South Korea). These porous media are widely employed in green roof systems. Total average diameter and density was 2.2 mm and 0.10e0.12, respectively, including 0.2mm perlite as growth type. For batches C and D, runoff quantity was measured with a 1-L graduated cylinder every hour. 2.2. Data and statistical analysis For 7 rainfall events, the collected discrete samples were analyzed in the laboratory at Seoul National University. The pH, EC, and turbidity were measured by 550A model (Orion, USA) and 2100P model (HACH, USA). We performed a correlation analysis to determine the possible relationship among Qpk/Qth.pk, the delayed occurrence time, and the antecedent dry day parameters. A nonparametric Spearman rank correlation was performed to compare relevant parameters, using SPSS v. 12.0 K software (SPSS, Inc., USA).
Table 2 Rainfall characteristics. Rainfall events (2003)
Rainfall duration
E1
5.07. 04:00e5.07. 06:00 5.31. 07:00e5.31. 24:00 6.22. 08:00e6.22. 24:00 7.03. 04:00e7.03. 19:00 7.07. 04:00e7.07. 21:00 8.30. 16:00e8.30. 19:00 9.29. 05:00e9.29. 13:00
E2 E3 E4 E5 E6 E7
Date. BeginningeEnd
Sustainment time
Total rainfall (mm)
3. Result and discussion 3.1. Rainwater runoff quantity characteristic of green roof
2h
3.5
17 h
7.5
14 h
16
15 h
115.0
17 h
42.5
3h
8.5
8h
22.5
There was no runoff from batches C and D in the tests with low rainfall in rainfall events E1 and E2. The data in Table 3 and Fig. 3 provide the peak flow and time of occurrence of the peak flow from the green roof for rainfalls E3eE7. The elapsed time means the time since the start of the rainfall. Theoretical runoff was calculated on multiplying hourly rainfall by the catchment area (1 m2). This value was similar to the runoff from batch test B with the concrete catchment, which was over 95%. It was considered that the error of less than 5% was due to water being absorbed by the concrete or lost to evaporation. Theoretical and actual rainwater runoff in representative rainfall
J.Y. Lee et al. / Journal of Environmental Management 152 (2015) 171e176
173
Fig. 2. Experimental Set-up. (a) Experiment A, (b) Experiment B, (c) Experiments C and D and cross section of test batches. (d) Norton rainfall simulator.
events had similar patterns. Theoretical rainwater runoff in rainfall event E3 had a peak flow (Qth.pk) of 8.55 L/h after an elapsed time of 11 h (June 22 at 5 p.m.) as indicated by the data in Table 3 and Fig. 3. However, batch tests C and D in event E3 had 1.57 L/h peak flow (Qpk) and 0.74 L/h after 14 h (June 22 at 8 p.m.), respectively. For reducing peak flow and delaying occurrence time, batches C and D reduced peak flow (Qpk/Qth.pk) to 0.18 and 0.09, respectively, and delayed occurrence time for 3 h. For rainfall event E5, theoretical rainwater runoff had a peak flow of 15.2 L/h after an elapsed time 14 h (July 7 at 5 p.m.) and actual rainwater runoff had a peak flow of 3.35 L/h after an elapsed time 17 h (July 7 at 5 p.m.) for batch test C and 1.98 L/h after an elapsed time 18 h (July 7 at 6 p.m.) for batch test D. Also, the different soil depth influenced the delay of the peak flow time. The peak flow reduction was 0.22 for the C test and 0.13 for the D test. For event E6, the theoretical runoff had peak flow of 6.37 L/h after an elapsed time 2 h (August 30 at 5 p.m.). Batch tests C and D had the peak flows of 1.32 L/h and 1.08 L/h, respectively, after 4 h (August 30 at 7 p.m.). The peak flow was reduced to 0.21 in the C test and 0.17 in the D test with a 2-h delay in the occurrence time. Event E7 had a peak flow of 1.93 L/h after an elapsed time 6 h in the C test and 1.34 L/h after an elapsed time 7 h in the D test after a rainfall event. The peak flow was reduced to 0.29 in the C test and 0.20 in the D test. Obviously, the soil depth is an important factor for reducing peak flow in the green roof system. Reducing peak flow in the green roof system indicates the amount of water that a
soil can store. Soil porosity and texture have a significant effect for that. In many respects, the soil porosity and texture in the green roof system serve an effective purpose by retaining water and delaying peak flow as it passes from the soil media to drainage system. The soil media act as a series of pipelines. These many different pipes store water and control the infiltration rate at which
Table 3 Runoff from a green roof. Events Duration (time)
Theoretical max runoff
Method Runoff amount Ratio (L/hr)
E3
Tpeak ¼ 11 h (6.22.17:00) Qpeak ¼ 8.55 L/h
C
E5
E6
E7
07:00 e24:00
04:00 e21:00
16:00 e19:00
05:00 e13:00
Tpeak ¼ 14 h (7.07.17:00) Qpeak ¼ 15.2 L/h Tpeak ¼ 2 h (8.30.17:00) Qpeak ¼ 6.37 L/h Tpeak ¼ 4 h (9.29.08:00) Qpeak ¼ 6.65 L/h
D C D C D C D
Tpeak ¼ 14 h Qpeak ¼ 1.57 L/h Tpeak ¼ 14 h Qpeak ¼ 0.74 L/h Tpeak ¼ 17 h Qpeak ¼ 3.35 L/h Tpeak ¼ 18 h Qpeak ¼ 1.98 L/h Tpeak ¼ 4 h Qpeak ¼ 1.32 L/h Tpeak ¼ 4 h Qpeak ¼ 1.08 L/h Tpeak ¼ 6 h Qpeak ¼ 1.93 L/h Tpeak ¼ 7 h Qpeak ¼ 1.34 L/h
Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼
0.18 0.09 0.22 0.13 0.21 0.17 0.29 0.20
174
J.Y. Lee et al. / Journal of Environmental Management 152 (2015) 171e176
Fig. 3. Theoretical rainwater runoff quantity and comparison with runoff quantity of test batches C and D.
flows through the soil media. However, the delayed occurrence time was not effective in the C and D tests, because perlite has high permeability and prevents soil compaction. In addition, the experimental soil depth conditions (i.e., 100 mm in C and 150 mm in D) were limited for identification of the delayed occurrence time. The data in Table 4 show that the correlation among the Qpk/Qth.pk, the delayed occurrence time, and the antecedent dry day (ADD) had a significant relationship through correlation coefficient (g). The data in Table 4 show that delayed occurrence time and ADD had a significantly positive value of 0.855 (p < 0.01). The ADD had a stronger relationship with the delayed occurrence time (g ¼ 0.855, p < 0.01). The ADD is an important factor for water retaining
Table 5 Runoff reduction effect by green roof. Rainfall event
Rainfall characteristic
Test Total runoff [L]
E3
Total rainfall: 16 mm Theoretical runoff: 15.2 L Duration of Rainfall: 14 h Total rainfall: 42.5 mm Theoretical runoff: 40.4 L Duration of Rainfall: 17 h Total rainfall: 8.5 mm Theoretical runoff: 8.1 L Duration of Rainfall: 3 h Total rainfall: 22.5 mm Theoretical runoff: 21.4 L Duration of Rainfall: 8 h
C
9.97
5.23
34.4
D
5.96
9.24
60.8
C
29.77
10.63
26.3
D
23.12
17.28
42.8
C
6.98
1.12
13.8
D
4.49
3.61
44.6
C
14.16
7.24
33.8
D
11.25
10.15
47.4
E5
E6 Table 4 Spearman correlation coefficients among the Qpk/Qth.pk, the delayed occurrence time, and the antecedent dry days (ADD) (4 events).
Qpk/Qth.pk Delayed occurrence time Antecedent dry days Qactual
Qpk/ Qth.pk
Delayed occurrence time
Antecedent dry days
1.00 0.583
0.583 1.00
0.295 0.855**
0.295
peak flow(Qpk)/Qtheoretical peak flow
0.855** (Qth.pk); *p < 0.05;
1.00 **
p < 0.01.
E7
Reduction of runoff
Ratio of reduction (%)
J.Y. Lee et al. / Journal of Environmental Management 152 (2015) 171e176
175
Fig. 4. Cumulative runoff quantity curve.
capacity. Increasing ADD represents the possibility for improving water retaining capacity. Qpk/Qth.pk has a negative relationship to the delayed occurrence time (g ¼ 0.583) and ADD (g ¼ 0.583). A high Qpk/Qth.pk means that the actual peak flow approaches the theoretical peak flow. Interestingly, high rainfall intensity has a negative relationship with the delayed occurrence time, demonstrating that the soil media in an extensive green roof do not efficiently retain rainwater, whereas the ADD has a weak negative relationship with Qpk/Qth.pk. The water holding capacity is decreased in high Qpk/Qth.pk level. That means the flood peak time is not effectively extended by the extensive green roof. The data in Table 5 and Fig. 4 provide the runoff reduction efficiency and the cumulative runoff quantity for 4 rainfall events (E3, E5, E6, and E7). In E3 the ratios of reduction were 34.4% in batch C and 60.8% in batch D. In E5, the ratios of reduction were 26.3% in C and 42.8% in D. In E6 the ratios of reduction were 13.8% and 44.6% in C and D, respectively. In E7, the ratios of reduction were 33.8% and 47.4%, respectively. In batches C and D, the average ratios of runoff reduction were about 27% and 49%, respectively. Judging by that, the runoff reduction effect was improved by increasing soil depth. The soil depth was an important factor for green roof design. As the soil depth increased, the runoff quantity decreased, and runoff time increased. However, the building roof could not carry the weight of
the deeper soil.
4. Summary and conclusion This research was performed to investigate the effects of reducing runoff through an extensive green roof based on 4 types of rainfall events using a rainfall simulator. 1) The extensive green roof system was analyzed, and it was found that it achieved a 13.8e60.8% reduction in runoff for the total rainfall in the test period by retention and the evapotranspiration effect from the soil layer and vegetation layer. Particularly, there was a 42.8e60.8% reduction in runoff with 200 mm soil depth and a 13.8e34.4% reduction in runoff with 150 mm soil depth. 2) In a Spearman rank correlation analysis, the delayed occurrence time and antecedent dry day (ADD) had a significantly positive value. Increasing ADD represents the possibility for improving water retention capacity and delayed occurrence time. Interestingly, high rainfall intensity had a negative relationship with delayed occurrence time, demonstrating that the soil media in an extensive green roof do not efficiently retain rainwater, whereas the ADD has a weak negative relationship with Qpk/
176
J.Y. Lee et al. / Journal of Environmental Management 152 (2015) 171e176
Qth.pk. A high Qpk/Qth.pk value means that the water-holding capacity was low, demonstrating that a high Qpk/Qth.pk value in an extensive green roof did not effectively extend the flood peak time. In many respects, the green roof systems serve a very effective purpose by runoff control. The green roof system acts as both retaining water for reducing peak flow. For retaining water, the soil texture and porosity act as a series of pipelines. These many different pipes store water and control the flow rate at which water flows through the soil media. In addition, this study showed that green roofs are highly effective for small-magnitude rainfall events. Acknowledgments This research was funded by KIST (Korea Institute of Science and Technology)-Natural Products Research Center (#2Z04211 (2Z04371) and 2Z04223 (2Z04383)). The authors would like to thank members for their help and involvement in the completion of the project. References Ayata, T., Tabares-Velasco, P.C., Srebric, J., 2011. An investigation of sensible heat fluxes at a green roof in a laboratory setup. Build. Environ. 46, 1851e1861.
Bengtsson, L., Grahn, L., Olsson, J., 2005. Hydrological function of a thin extensive green roof in southern Sweden. Nord. Hydrol. 36, 259e268. Berndtsson, J.C., 2010. Green roof performance towards management of runoff water quantity and quality: a review. Ecol. Eng. 36, 351e360. Berndtsson, J.C., Bengtsson, L., Jinno, K., 2008. First flush effect from vegetated roofs during simulated rain events. Hydrol. Res. 39, 171e179. Berndtsson, J.C., Emilsson, T., Bengtsson, L., 2006. The influence of extensive vegetated roofs on runoff water quality. Sci. Total Environ. 355, 369e380. Del Barrio, E.P., 1998. Analysis of green roofs cooling potential in buildings. Energy Build. 27, 179e193. Emilsson, T., Berndtsson, J.C., Mattsson, J.E., Rolf, K., 2007. Effect of using conventional and controlled release fertiliser on nutrient runoff from various vegetated roof systems. Ecol. Eng. 29, 260e271. Mentens, J., Raes, D., Hermy, M., 2006. Green roofs as a tool for solving the rainwater runoff problem in the urbanized 21st century. Landsc. Urban Plan. 77, 217e226. Moran, A., Hunt, B., Jennings, G., 2003. A north Carolina field study to evaluate green-roof runoff quantity, runoff quality, and plant growth. ASCE Conf. Proc. 118, 335. Nagase, A., Dunnett, N., 2010. Drought tolerance in different vegetation types for extensive green roofs: effects of watering and diversity. Landsc. Urban Plan. 97, 318e327. Tabares-Velasco, P.C., Srebric, J., 2009. The role of plants in the reduction of heat flux through green roofs: laboratory experiments. ASHRAE Trans. 2009, 115. Teemusk, A., Mander, U., 2007. Rainwater runoff quantity and quality performance from a green roof: the effects of short term rain events. Ecol. Eng. 30, 271e277. Vijayaraghavan, K., Joshi, U.M., Balasubramanian, R., 2012. A field study to evaluate runoff quality from green roofs. Water Res. 46, 1337e1345. Villarreal, E.L., Bengtsson, L., 2005. Response of a Sedum green roof to individual rain events. Ecol. Eng. 25, 1e7. Wong, N.H., Chen, Y., Ong, C.L., Sia, A., 2003. Investigation of thermal benefits of rooftop garden in the tropical environment. Build. Environ. 38, 261e270.
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
A pilot study to evaluate runoff quantity from green roofs Ju Young Lee a, Min Jung Lee c, Mooyoung Han b, * a
KIST(Korea Institute of Science and Technology)-Natural Products Research Center, Gangnueng, 210-340, South Korea Department of Civil and Environmental Engineering, Seoul National University, 599 Gwanak-ro, Gwanak-gu, Seoul, 151-744, South Korea c LIFTRC(LED-IT Fusion Technology Research Center), Daegu, South Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 January 2013 Received in revised form 24 July 2014 Accepted 17 January 2015 Available online
The use of green roofs is gaining increased recognition in many countries as a solution that can be used to improve environmental quality and reduce runoff quantity. To achieve these goals, pilot-scale green roof assemblies have been constructed and operated in an urban setting. From a stormwater management perspective, green roofs are 42.8e60.8% effective in reducing runoff for 200 mm soil depth and 13.8 e34.4% effective in reducing runoff for 150 mm soil depth. By using Spearman rank correlation analysis, high rainfall intensity was shown to have a negative relationship with delayed occurrence time, demonstrating that the soil media in green roofs do not efficiently retain rainwater. Increasing the number of antecedent dry days can help to improve water retention capacity and delay occurrence time. From the viewpoint of runoff water quality, green roofs are regarded as the best management practice by filtration and adsorption through growth media (soil). © 2015 Elsevier Ltd. All rights reserved.
Keywords: Green roofs Environmental quality Reducing runoff quantity Spearman rank correlation Best management practice (BMP)
1. Introduction Recently, green roofs have been receiving increased recognition in many countries such as the USA, Japan, and European countries (Vijayaraghavan et al., 2012; Berndtsson et al., 2008; Mentens et al., 2006). Particularly, the installation of green roofs is now increasing in Korea, as shown Fig. 1. Green roofs have several positive effects in the urban setting. The most important effect is their ability to retain and detain rainwater (Villarreal and Bengtsson, 2005). Green roofs have been introduced for reducing urban stormwater runoff from rooftops. The growth media in the green roof system typically retains and detains rainwater. The depth of the growth media can play an important role when used in an eco-friendly drainage system by slowing and reducing runoff volume. Retaining rainwater in soil reduces peak flow, which means that it prolongs the time of concentration (Bengtsson et al., 2005). The green roof has several benefits that include reduction of the urban heat island effect (Wong et al., 2003) and reducing building energy consumption by cooling roofs during the summer (Vijayaraghavan et al., 2012; Del Barrio, 1998). Finally, green roofs improve urban biodiversity and create habitat for plants and animals in addition to their
* Corresponding author. E-mail addresses: [email protected] (J.Y. Lee), [email protected] (M.J. Lee), [email protected] (M. Han). http://dx.doi.org/10.1016/j.jenvman.2015.01.028 0301-4797/© 2015 Elsevier Ltd. All rights reserved.
esthetic appeal (Vijayaraghavan et al., 2012; Emilsson et al., 2007). Green roofs are mainly divided into two types. Intensive green roofs are characterized by a thick layer of growth medium of more than 200 mm thickness, and extensive green roofs have a layer of less than 200 mm thickness (Nagase and Dunnett, 2010). Most green roof systems consist of a drainage layer, a root barrier, and a waterproof membrane. Previous studies have focused on evaporative and radiative heat transfer mechanisms of the different types of roof assemblies (Ayata et al., 2011; Tabares-Velasco and Srebric, 2009). Many previous studies have considered the runoff quality from green roofs (Ayata et al., 2011; Berndtsson et al., 2008, 2006). These studies show that green roofs can filter and absorb nonpoint pollutants (Berndtsson et al., 2008, 2006). However, in the longterm they can contribute to the deterioration of runoff water quality by releasing fertilizers regarded as pollutants (Berndtsson, 2010; Teemusk and Mander, 2007; Berndtsson et al., 2006). Moran et al. (2003) found that green roof systems have advantages and disadvantages. They are the best management practice for achieving water retention and peak flow reduction as benefits; however, they result in high nutrient concentrations. Thus, the main purpose of the present paper is to evaluate the reduction of peak flow, which prolongs the time of concentration during rainfall events.
172
J.Y. Lee et al. / Journal of Environmental Management 152 (2015) 171e176
Fig. 1. Extensive green roof (NAM-SAN annex, Seoul City Hall, Korea). Table 1 Outline of each test batch. Material
A B C D
Size (cm)
Acryl Concrete Green roof
Width
Length
Height
70 100 100
70 100 100
e 3 15 20
Soil
Vegetation
Purpose
Measure list
Note
e e Perlite
e e Sedum
Rainfall sampling Runoff sampling Green roof outflow
W$Qual, W$Quan W$Qual, W$Quan
Fig. 1(a) Fig. 1(b) Fig. 1(c)
W$Qual: Water Quality; W$Quan: Water Quantity.
2. Methods and materials 2.1. Data collection The quantity and quality of runoff were evaluated for 4 types of pilot facilities that used acryl, concrete, and 2 different green roof models, and were established on the rooftop of Seoul National University Building 35, Gwanak-Gu, Seoul, Korea (see Table 1). There were 7 rainfall events in the test period from May 7, 2011 to Sept 29, 2011. Total rainfall data and rainfall intensities were obtained for Seoul from the Seoul Meteorological Office (see Table 2). A Norton Rainfall Simulator with 2 heads, model (DIK-600), which was developed by USDA-ARS NSERL at Purdue University, West Lafayette, Indiana, was used to simulate the 7 rainfall events. For harvesting rainwater unaffected by the catchment area, the catchment facility was made of acrylic material (Fig. 2(a)). The catchment installed in test batch B was made of concrete to examine the effect of the catchment on runoff quality (Fig. 2(b)). In Fig. 2(c), batches C (150 mm) and D (200 mm) were sampled at different soil depths to examine the relationship of runoff quantity
and water retention with soil depth. Further, Fig. 2(c) shows a cross section of the test batches C and D. These test batches were composed of layers of sedum, volcanic materials and soil with peat moss (50 mm), perlite (100 mm for test C and 150 mm for test D), and a drainage plate (40e50 mm). The soils were provided by the producer (GreenInfra, Co., Ltd., South Korea). These porous media are widely employed in green roof systems. Total average diameter and density was 2.2 mm and 0.10e0.12, respectively, including 0.2mm perlite as growth type. For batches C and D, runoff quantity was measured with a 1-L graduated cylinder every hour. 2.2. Data and statistical analysis For 7 rainfall events, the collected discrete samples were analyzed in the laboratory at Seoul National University. The pH, EC, and turbidity were measured by 550A model (Orion, USA) and 2100P model (HACH, USA). We performed a correlation analysis to determine the possible relationship among Qpk/Qth.pk, the delayed occurrence time, and the antecedent dry day parameters. A nonparametric Spearman rank correlation was performed to compare relevant parameters, using SPSS v. 12.0 K software (SPSS, Inc., USA).
Table 2 Rainfall characteristics. Rainfall events (2003)
Rainfall duration
E1
5.07. 04:00e5.07. 06:00 5.31. 07:00e5.31. 24:00 6.22. 08:00e6.22. 24:00 7.03. 04:00e7.03. 19:00 7.07. 04:00e7.07. 21:00 8.30. 16:00e8.30. 19:00 9.29. 05:00e9.29. 13:00
E2 E3 E4 E5 E6 E7
Date. BeginningeEnd
Sustainment time
Total rainfall (mm)
3. Result and discussion 3.1. Rainwater runoff quantity characteristic of green roof
2h
3.5
17 h
7.5
14 h
16
15 h
115.0
17 h
42.5
3h
8.5
8h
22.5
There was no runoff from batches C and D in the tests with low rainfall in rainfall events E1 and E2. The data in Table 3 and Fig. 3 provide the peak flow and time of occurrence of the peak flow from the green roof for rainfalls E3eE7. The elapsed time means the time since the start of the rainfall. Theoretical runoff was calculated on multiplying hourly rainfall by the catchment area (1 m2). This value was similar to the runoff from batch test B with the concrete catchment, which was over 95%. It was considered that the error of less than 5% was due to water being absorbed by the concrete or lost to evaporation. Theoretical and actual rainwater runoff in representative rainfall
J.Y. Lee et al. / Journal of Environmental Management 152 (2015) 171e176
173
Fig. 2. Experimental Set-up. (a) Experiment A, (b) Experiment B, (c) Experiments C and D and cross section of test batches. (d) Norton rainfall simulator.
events had similar patterns. Theoretical rainwater runoff in rainfall event E3 had a peak flow (Qth.pk) of 8.55 L/h after an elapsed time of 11 h (June 22 at 5 p.m.) as indicated by the data in Table 3 and Fig. 3. However, batch tests C and D in event E3 had 1.57 L/h peak flow (Qpk) and 0.74 L/h after 14 h (June 22 at 8 p.m.), respectively. For reducing peak flow and delaying occurrence time, batches C and D reduced peak flow (Qpk/Qth.pk) to 0.18 and 0.09, respectively, and delayed occurrence time for 3 h. For rainfall event E5, theoretical rainwater runoff had a peak flow of 15.2 L/h after an elapsed time 14 h (July 7 at 5 p.m.) and actual rainwater runoff had a peak flow of 3.35 L/h after an elapsed time 17 h (July 7 at 5 p.m.) for batch test C and 1.98 L/h after an elapsed time 18 h (July 7 at 6 p.m.) for batch test D. Also, the different soil depth influenced the delay of the peak flow time. The peak flow reduction was 0.22 for the C test and 0.13 for the D test. For event E6, the theoretical runoff had peak flow of 6.37 L/h after an elapsed time 2 h (August 30 at 5 p.m.). Batch tests C and D had the peak flows of 1.32 L/h and 1.08 L/h, respectively, after 4 h (August 30 at 7 p.m.). The peak flow was reduced to 0.21 in the C test and 0.17 in the D test with a 2-h delay in the occurrence time. Event E7 had a peak flow of 1.93 L/h after an elapsed time 6 h in the C test and 1.34 L/h after an elapsed time 7 h in the D test after a rainfall event. The peak flow was reduced to 0.29 in the C test and 0.20 in the D test. Obviously, the soil depth is an important factor for reducing peak flow in the green roof system. Reducing peak flow in the green roof system indicates the amount of water that a
soil can store. Soil porosity and texture have a significant effect for that. In many respects, the soil porosity and texture in the green roof system serve an effective purpose by retaining water and delaying peak flow as it passes from the soil media to drainage system. The soil media act as a series of pipelines. These many different pipes store water and control the infiltration rate at which
Table 3 Runoff from a green roof. Events Duration (time)
Theoretical max runoff
Method Runoff amount Ratio (L/hr)
E3
Tpeak ¼ 11 h (6.22.17:00) Qpeak ¼ 8.55 L/h
C
E5
E6
E7
07:00 e24:00
04:00 e21:00
16:00 e19:00
05:00 e13:00
Tpeak ¼ 14 h (7.07.17:00) Qpeak ¼ 15.2 L/h Tpeak ¼ 2 h (8.30.17:00) Qpeak ¼ 6.37 L/h Tpeak ¼ 4 h (9.29.08:00) Qpeak ¼ 6.65 L/h
D C D C D C D
Tpeak ¼ 14 h Qpeak ¼ 1.57 L/h Tpeak ¼ 14 h Qpeak ¼ 0.74 L/h Tpeak ¼ 17 h Qpeak ¼ 3.35 L/h Tpeak ¼ 18 h Qpeak ¼ 1.98 L/h Tpeak ¼ 4 h Qpeak ¼ 1.32 L/h Tpeak ¼ 4 h Qpeak ¼ 1.08 L/h Tpeak ¼ 6 h Qpeak ¼ 1.93 L/h Tpeak ¼ 7 h Qpeak ¼ 1.34 L/h
Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼ Qpk.C/ Qth.pk ¼
0.18 0.09 0.22 0.13 0.21 0.17 0.29 0.20
174
J.Y. Lee et al. / Journal of Environmental Management 152 (2015) 171e176
Fig. 3. Theoretical rainwater runoff quantity and comparison with runoff quantity of test batches C and D.
flows through the soil media. However, the delayed occurrence time was not effective in the C and D tests, because perlite has high permeability and prevents soil compaction. In addition, the experimental soil depth conditions (i.e., 100 mm in C and 150 mm in D) were limited for identification of the delayed occurrence time. The data in Table 4 show that the correlation among the Qpk/Qth.pk, the delayed occurrence time, and the antecedent dry day (ADD) had a significant relationship through correlation coefficient (g). The data in Table 4 show that delayed occurrence time and ADD had a significantly positive value of 0.855 (p < 0.01). The ADD had a stronger relationship with the delayed occurrence time (g ¼ 0.855, p < 0.01). The ADD is an important factor for water retaining
Table 5 Runoff reduction effect by green roof. Rainfall event
Rainfall characteristic
Test Total runoff [L]
E3
Total rainfall: 16 mm Theoretical runoff: 15.2 L Duration of Rainfall: 14 h Total rainfall: 42.5 mm Theoretical runoff: 40.4 L Duration of Rainfall: 17 h Total rainfall: 8.5 mm Theoretical runoff: 8.1 L Duration of Rainfall: 3 h Total rainfall: 22.5 mm Theoretical runoff: 21.4 L Duration of Rainfall: 8 h
C
9.97
5.23
34.4
D
5.96
9.24
60.8
C
29.77
10.63
26.3
D
23.12
17.28
42.8
C
6.98
1.12
13.8
D
4.49
3.61
44.6
C
14.16
7.24
33.8
D
11.25
10.15
47.4
E5
E6 Table 4 Spearman correlation coefficients among the Qpk/Qth.pk, the delayed occurrence time, and the antecedent dry days (ADD) (4 events).
Qpk/Qth.pk Delayed occurrence time Antecedent dry days Qactual
Qpk/ Qth.pk
Delayed occurrence time
Antecedent dry days
1.00 0.583
0.583 1.00
0.295 0.855**
0.295
peak flow(Qpk)/Qtheoretical peak flow
0.855** (Qth.pk); *p < 0.05;
1.00 **
p < 0.01.
E7
Reduction of runoff
Ratio of reduction (%)
J.Y. Lee et al. / Journal of Environmental Management 152 (2015) 171e176
175
Fig. 4. Cumulative runoff quantity curve.
capacity. Increasing ADD represents the possibility for improving water retaining capacity. Qpk/Qth.pk has a negative relationship to the delayed occurrence time (g ¼ 0.583) and ADD (g ¼ 0.583). A high Qpk/Qth.pk means that the actual peak flow approaches the theoretical peak flow. Interestingly, high rainfall intensity has a negative relationship with the delayed occurrence time, demonstrating that the soil media in an extensive green roof do not efficiently retain rainwater, whereas the ADD has a weak negative relationship with Qpk/Qth.pk. The water holding capacity is decreased in high Qpk/Qth.pk level. That means the flood peak time is not effectively extended by the extensive green roof. The data in Table 5 and Fig. 4 provide the runoff reduction efficiency and the cumulative runoff quantity for 4 rainfall events (E3, E5, E6, and E7). In E3 the ratios of reduction were 34.4% in batch C and 60.8% in batch D. In E5, the ratios of reduction were 26.3% in C and 42.8% in D. In E6 the ratios of reduction were 13.8% and 44.6% in C and D, respectively. In E7, the ratios of reduction were 33.8% and 47.4%, respectively. In batches C and D, the average ratios of runoff reduction were about 27% and 49%, respectively. Judging by that, the runoff reduction effect was improved by increasing soil depth. The soil depth was an important factor for green roof design. As the soil depth increased, the runoff quantity decreased, and runoff time increased. However, the building roof could not carry the weight of
the deeper soil.
4. Summary and conclusion This research was performed to investigate the effects of reducing runoff through an extensive green roof based on 4 types of rainfall events using a rainfall simulator. 1) The extensive green roof system was analyzed, and it was found that it achieved a 13.8e60.8% reduction in runoff for the total rainfall in the test period by retention and the evapotranspiration effect from the soil layer and vegetation layer. Particularly, there was a 42.8e60.8% reduction in runoff with 200 mm soil depth and a 13.8e34.4% reduction in runoff with 150 mm soil depth. 2) In a Spearman rank correlation analysis, the delayed occurrence time and antecedent dry day (ADD) had a significantly positive value. Increasing ADD represents the possibility for improving water retention capacity and delayed occurrence time. Interestingly, high rainfall intensity had a negative relationship with delayed occurrence time, demonstrating that the soil media in an extensive green roof do not efficiently retain rainwater, whereas the ADD has a weak negative relationship with Qpk/
176
J.Y. Lee et al. / Journal of Environmental Management 152 (2015) 171e176
Qth.pk. A high Qpk/Qth.pk value means that the water-holding capacity was low, demonstrating that a high Qpk/Qth.pk value in an extensive green roof did not effectively extend the flood peak time. In many respects, the green roof systems serve a very effective purpose by runoff control. The green roof system acts as both retaining water for reducing peak flow. For retaining water, the soil texture and porosity act as a series of pipelines. These many different pipes store water and control the flow rate at which water flows through the soil media. In addition, this study showed that green roofs are highly effective for small-magnitude rainfall events. Acknowledgments This research was funded by KIST (Korea Institute of Science and Technology)-Natural Products Research Center (#2Z04211 (2Z04371) and 2Z04223 (2Z04383)). The authors would like to thank members for their help and involvement in the completion of the project. References Ayata, T., Tabares-Velasco, P.C., Srebric, J., 2011. An investigation of sensible heat fluxes at a green roof in a laboratory setup. Build. Environ. 46, 1851e1861.
Bengtsson, L., Grahn, L., Olsson, J., 2005. Hydrological function of a thin extensive green roof in southern Sweden. Nord. Hydrol. 36, 259e268. Berndtsson, J.C., 2010. Green roof performance towards management of runoff water quantity and quality: a review. Ecol. Eng. 36, 351e360. Berndtsson, J.C., Bengtsson, L., Jinno, K., 2008. First flush effect from vegetated roofs during simulated rain events. Hydrol. Res. 39, 171e179. Berndtsson, J.C., Emilsson, T., Bengtsson, L., 2006. The influence of extensive vegetated roofs on runoff water quality. Sci. Total Environ. 355, 369e380. Del Barrio, E.P., 1998. Analysis of green roofs cooling potential in buildings. Energy Build. 27, 179e193. Emilsson, T., Berndtsson, J.C., Mattsson, J.E., Rolf, K., 2007. Effect of using conventional and controlled release fertiliser on nutrient runoff from various vegetated roof systems. Ecol. Eng. 29, 260e271. Mentens, J., Raes, D., Hermy, M., 2006. Green roofs as a tool for solving the rainwater runoff problem in the urbanized 21st century. Landsc. Urban Plan. 77, 217e226. Moran, A., Hunt, B., Jennings, G., 2003. A north Carolina field study to evaluate green-roof runoff quantity, runoff quality, and plant growth. ASCE Conf. Proc. 118, 335. Nagase, A., Dunnett, N., 2010. Drought tolerance in different vegetation types for extensive green roofs: effects of watering and diversity. Landsc. Urban Plan. 97, 318e327. Tabares-Velasco, P.C., Srebric, J., 2009. The role of plants in the reduction of heat flux through green roofs: laboratory experiments. ASHRAE Trans. 2009, 115. Teemusk, A., Mander, U., 2007. Rainwater runoff quantity and quality performance from a green roof: the effects of short term rain events. Ecol. Eng. 30, 271e277. Vijayaraghavan, K., Joshi, U.M., Balasubramanian, R., 2012. A field study to evaluate runoff quality from green roofs. Water Res. 46, 1337e1345. Villarreal, E.L., Bengtsson, L., 2005. Response of a Sedum green roof to individual rain events. Ecol. Eng. 25, 1e7. Wong, N.H., Chen, Y., Ong, C.L., Sia, A., 2003. Investigation of thermal benefits of rooftop garden in the tropical environment. Build. Environ. 38, 261e270.
