Beyond Drainage: The role of SUDS in the mitigation of Urban

Heat Island effects

S. L. Moore*, E. A. Hathway**, M. Fahmy*** and V. R. Stovin**


*URSULA project, University of Sheffield, ICOSS, 219 Portobello, Sheffield, S. Yorkshire S1 4DP, UK
(E-mail: s.moore@sheffield.ac.uk)
**Department of Civil and Structural Engineering, University of Sheffield, Mappin Street, Sheffield S1 3JD, UK
(E-mail: a.hathway@sheffield.ac.uk, v.stovin@sheffield.ac.uk)
***Department of Architecture, Military Technical College, Cairo, Egypt (E-mail: md.fahmy@live.com)

Abstract
In order to understand the interplay between SUDS design and the effects on microclimate, two
distinctive, hypothetical, urban designs have been created for a heavily urbanised UK case study
area with a riverside location. The two hypothetical case-study designs have been analysed for
their hydraulic and water quality performance, and the impact on microclimate using the
modelling software tools MUSIC and ENVI-met respectively. The results of this work show that
urban greening through the use of SUDS has positive impacts not only on hydrology, but also on
local air temperatures. The selection of, and the layout of, SUDS devices within the urban area
also has an effect on both the hydraulic performance and the surrounding microclimate. This has
implications for urban planning and site design.

Keywords
Sustainable Drainage Systems, Urban Heat Island, Green Infrastructure.


INTRODUCTION
The effect of urbanisation on both air temperatures and hydrology are well described (e.g.
Biesbroek 2010). Urbanisation replaces vegetated surfaces which provide shading, evaporative
cooling and rainwater interception, storage and infiltration functions with impervious surfaces.
Urban greenspaces, such as those that can be created through the use of Green Infrastructure (GI)
provide areas within the built environment where these processes can take place, and therefore offer
potential to help adapt areas to the effects of climate change and mitigate against urban heat island
(UHI) effects.

[a] network of multi-functional green space, both new
and existing, both rural and urban, which supports the natural and ecological processes and is
integral to the health and quality of life of sustainable communities . GI comprises many forms; at
all scales from individual green roofs or urban parks, to large scale regional parks.

In the context of this paper, one potential manifestation of GI will be discussed, Sustainable
Drainage Systems (SUDS). SUDS is a generic term used in the UK to refer to various measures
aimed at controlling surface water runoff (and consequent flooding and pollution problems) from
urban catchments. Structural SUDS include green roofs, soakaways, swales, infiltration trenches
ponds and wetlands. Internationally SUDS are variously referred to as Best Management Practices,
Low Impact Design (LID) and Water Sensitive Urban Design (WSUD). Due to their reliance on
natural catchment processes, these technologies
underground pipe and storage-based solutions.

To date, most SUDS research has focussed on either the performance of SUDS structures (eg.
Scholes et al. 2005) or the effect of a number of SUDS in series (treatment trains) (e.g Bastien et al.
2010). Several studies have produced decision support tools to guide the user to select the most
appropriate SUDS for a range of implementation drivers (e.g. Cheng et al. 2009; Ellis et al. 2005;
Stovin and Swan, 2007). However, less attention has been paid to the relationship between SUDS
Manuscript
and Urban Design (notable exceptions include Stahre, 2008 and Echols, 2007). Or the effect that
SUDS selection and placement have on not only their hydraulic and pollutant removal performance,
but also on wider environmental and sustainability factors, such as microclimate. Despite the fact
these multiple benefits are routinely cited.

The urban microclimate is of significant importance to the health and wellbeing of city dwellers.
Even the simple removal of vegetation adjacent to a property can increase the risk of mortality
during heat waves (Vendentorren et al., 2006). Alongside vegetation removal, the increase in hard
impervious surfaces rapidly removes water from the immediate environment, preventing cooling by
-
wave radiation from escaping to the atmosphere, further heating up the locality (Grimmond and
Oke, 1999). There is much interest in reintroducing vegetation into urban areas to increase our cities
resilience to heat waves (Smith and Levermore, 2008). For instance, large parks have been shown to
provide significant levels of cooling in a variety of climates (e.g. Upmanis et al., 1998; Jauregi,
1990) yet how this propagates into the urban area depends on the local street design. There are also
indications that smaller, distributed, green spaces provide benefits for propagating the natural
cooling into the urban area (Shashua-bar and Hoffman, 2000). Methods for distributing green
coverage in semi arid regions is covered by Fahmy (2010), however there is, as yet, no conclusive
guidance on the most appropriate form of green infrastructure for UHI mitigation in Europe.
Guidance should be combined with an investigation of different distributions of SUDS measures to
provide integrated solutions for climate change resilience accounting for both heatwaves and rain
storms. This paper will focus on this question, quantifying the relative benefits of different designs
for water management and UHI mitigation. Furthermore it will focus on SUDS and show that the
choice of measures and the location of their deployment within the catchment can have implications
for their performance.


CASE STUDY SCENARIO CONCEPTUALISATION
The case study site is an 11.3 ha urban area within Sheffield, UK. The site is bounded to the west,
south and east by the River Don. Currently, the site is predominantly mixed use; there are many
plots that are vacant or underused. Away from the river channel and banks, there are few areas of
green-space, much of the site is impermeable. In addition, the site has experienced flooding in
recent history.

Two distinct hypothetical redevelopment scenarios for the site were developed by the URSULA
project (Figure 1). The aim of this process was to produce two highly contrasting designs that could
be tested for a range of broad sustainability criteria. In order to aid the comparison of each
scenario, a baseline scenario, representing the case study site prior to regeneration, is also
modelled. For the two scenarios, basic SUDS concepts were employed, therefore Natural drainage
pathways are utilised; source control structures were sought wherever possible in preference to
regional or offsite controls. Treatment trains were also employed where appropriate. Consideration
of the local microclimate was also incorporated into the design by influencing the height / width
ratios of the streets, stepped building heights next to open areas were included to reduce the risk of
high wind speeds and vegetation, particularly trees were incorporated when it was in keeping with
the design ethos.

Figure 1. Conceptualisation of the URSULA redevelopment scenarios. Urban forms designed
by L. Pattacini and the URSULA design team. Maps Crown Copyright/Digimap 2011. An
Ordnance Survey/EDINA supplied service.


METHODOLOGY
The methodology employed within this study can be divided into the following sections:
1. Physical characterisation of each scenarios: Description of the land occupied by each key
design feature, focusing primarily on the SUDS treatment trains
2. Hydrological: Evaluation of each SUDS scenario in terms of peak flow and total volume
reduction for a range of rainfall profiles
3. Microclimate: Evaluation of the local wind speed, temperature, humidity and Predicted
Mean Vote (PMV), a measure of pedestrian comfort. Both designs are based in the same
location, with identical climates, however in order to demonstrate the benefits of the systems
different weather conditions are considered for the analysis of the hydrology and urban heat
island.

1) Physical characterisation of each scenario
The physical characteristics of each scenario were analysed from 2-D CAD representations in the
Geographical Information System (GIS) software ArcView v9.3.

2) Hydrological evaluation
Surface water runoff for each SUDS scenario was modelled using MUSIC (Model for Urban
Stormwater Improvement Conceptualisation). MUSIC is a hydrological model coupled with a
water quality model, developed by Wong et al. (2006). Presented within this study are results for
water quantity assessment only. Hydrological routing between structures is provided by utilising
Muskingum Cunge routing parameters to simulate travel time (K) and level of attenuation ( ).

SUDS Design assumptions
For the purposes of hydrological modelling, the site was assumed to be 90% impervious,
representing the high level of hardstanding and built surfaces within the catchment. The SUDS
designs for each scenario have been translated into MUSIC. Surface areas of each device were
estimated directly from the GIS plans for each scenario. Where practical, 1 in 100 yr 60 min
rainfall depths for Sheffield (53.04 mm) were used to calculate the depths of each structure. If, due
to space constraints, or if the volumes required to completely contain the 1 in 100 yr storm were not
available, then SUDS design guidance (e.g. Woods-Ballard et al., 2007) was used to identify
suitable SUDS which would provide at least partial management benefits. For example, Scenario 1
utilises green roofs throughout the scheme. Extensive roofs have been designed to provide 5 mm
storage across their entire surface area (Berryman and Stovin, 2010). Intensive Green roofs have
been modelled as roof gardens, with 30 mm soil depth. It must be noted that in the case of scenario
2, flows pertaining to the flood channel itself are not included as this would require interaction with
river flows, which is beyond the scope of this paper. A representative Greenfield runoff rate for the
site was assumed to equal 7 l/s/ha, which equates to 0.08 m
3
s for the 11.3 ha site.

3) Microclimate evaluation
The local microclimate within each scenario was modelled using ENVI-met BETA4 (Bruse, 1999),
a freeware numerical model that combines simulations of wind flow, radiant, sensible and latent
heat flux, alongside transpiration and evaporation from vegetation and soils. The influence of
vegetation on shading and the drag effects on air flow are included.

Microclimate modelling assumptions
In this model trees are incorporated following the method by Fahmy (2010) and assumed to be
mature. A key approximation for this work is that water is represented via a fully saturated soil.
This means that convective flow within any water body is not accounted for and heat transfer is
based only on a placid body of water (e.g. a pond on a calm day). The model also does not take into
account any heat output from traffic, but heat loss from the buildings is included. The simulations
are built on a 2 m grid, which requires some approximation of the SUDS designs. Trees are
included in planters and the bioretention swale. The extensive green roof is represented by short
grass, the intensive by 1/3 grass, 1/3 shrubs, and 1/3 small trees. Previous studies have provided
some validation for radiant heat flux and relative humidity (Ali-Toudert, 2005; Fahmy et al., 2010).

The climatic input parameters were based on the case studies location in Sheffield, UK (Lat; 53.38,
Long; -1.46). A transient simulation was run for 12 hours (8 am to 8 pm) using design data for a hot
day (27
th
July). Averages of the climatic variables are calculated based on Fahmy (2010)
and temperature profiles for the site presented.


RESULTS AND DISCUSSION
1) Physical characterisation
The levels of impervious area in both of the redevelopment scenarios are lower than the As Is
scenario (Table 1). Scenario 1 requires a smaller land take than Scenario 2. However, Scenario 2
provides a larger SUDS treatment train length.



Table 1. SUDS characteristics and land take associated with each redevelopment scenario.

Impervious
area drained
via SUDS
Max SUDS treatment
train length
% Number of units in
series
As- Is 0
-
41 2
47
4

2) Hydrological Evaluation
The results of the MUSIC hydrological simulations are presented in Figure 2. Runoff profiles are
presented for four representative storms. Two design storms have been modelled, the 30 yr 60 min
and 100 yr 60 min for Sheffield have been chosen as these storms are routinely used within
drainage design to test compliance with runoff legislation, and for exceedance flows. Two sampled
rainfall events are also presented, the 04/10/08 storm equates to a 1.38 yr return period for
Sheffield, UK, while the 13/06/2007 storm equates to a 15.97 yr return period and was the precursor
storm to the Sheffield Floods in late June 2007.

Figure 2. Rainfall runoff profiles from the entire 11.3 ha catchment for Scenarios 1 and 2
compared to the baseline As Is scenario for a range of rainfall intensities.
The rainfall-runoff profiles in Figure 2 show that the provision of SUDS goes some way to reduce
both the total volume and flow rate from the case study site. Further analysis of the modelling
results show that Scenario 2 achieves both a greater reduction of total volume, and attenuation of
peak runoff than Scenario 1 when compared to the As Is runoff profile (Table 2). Within each
scenario, the performance is likely to provide only a partial improvement, as a large proportion of
the area in each scenario is not served by SUDS (Table 1). When the runoff from the area of each
catchment actually served by SUDS was analysed, it was found that peak flows were reduced below
the assumed Greenfield runoff rate for the site in all but the 1 in 100 yr 60 min storm for both
Scenarios, and the 13/06/2007 storm for Scenario 2 (not presented).

Table 2. Comparison of hydrological parameters for each design scenario
Complete site runoff

Area served by SUDS
Storm profile Total volume m
3

(% volume
reduction)
Peak Q, m
3
s
(% peak Q
reduction)
Total volume
m
3

Peak Q, m
3
s
As Is 30 yr 60 min 2842.49 (-) 0.83 - -
100 yr 60 min 5391.42 (-) 1.56 - -
04/10/2008 2758.66 (-) 0.12 - -
13/06/2007

10258.83 (-) 0.32 - -
Scenario 1 30 yr 60 min 2238.51 (21) 0.56 (32) 544.41 0.05
100 yr 60 min 4479.41 (17) 1.10 (30) 1266.84 0.17
04/10/2008 2084.01 (24) 0.09 (30) 439.85 0.01
13/06/2007

8572.36 (16) 0.24 (27) 2435.89 0.05
Scenario 2 30 yr 60 min 2047.15 (28) 0.47 (43) 440.19 0.01
100 yr 60 min 4282.20 (21) 0.93 (41) 1232.93 0.12
04/10/2008 1910.26 (31) 0.07(40) 350.75 0.01
13/06/2007 8429.44 (18) 0.26 (19) 2672.63 0.10

3) Microclimate evaluation
The aim of the designs is to provide mitigation of high temperatures, and this is achieved to some
extent in both Scenarios 1 and 2. Both sites show reduced air temperatures (Figure 3) to the
Scenario, including late in the day when the UHI is at its greatest. The relative humidity (RH) on
each site reduces for all as air temperature increases, in the evening all sites have an RH lower than
70% which can be deemed comfortable.

Figure 3. Average air temperature for each scenario

The contour plots of temperature in Figure 4 demonstrate the variation across the whole site. Both
the and Scenario 1 have hot spots reaching temperatures greater than 25C. However, the
area covered by this extreme is much larger in the . In Scenario 1 the inclusion of
vegetation along the top of the site seems to mitigate this, reducing the local temperature by
approximately 1C. Scenario 2 has a more even and cooler temperature profile. This may be due to
increased wind velocities through the park of between 1.5 and 2.4 m/s. Despite the cooler air
temperatures in Scenario 2, both development scenarios have a similar level of comfort using the
Percent Mean Vote method (Jendritzy et al., 2002) which accounts for air temperature, radiant
temperature, humidity and wind speed. This is explained by Scenario 2 being open to solar radiation
and increased wind speeds, both of which can increase discomfort. However, the wind speeds are
still lower that that around the ring road in the base case (2.6 m/s), indicating the benefits of trees in
reducing wind speeds as well as decreasing air temperatures. It should be noted that this study only
considers summer conditions and consideration of the effect of high wind speeds during autumn and
winter weather should also be taken into account.


Figure 4. Profiles of air temperature across the site at 4 pm, and a height of 1.5 m


CONCLUSIONS
The inclusion of SUDS within each urban redevelopment scenario goes some way to reduce the
peak flows and volumes generated from the catchment. However, the overall effect is reduced due
to the impervious areas that are not presently disconnected from the sewer. As with the SUDS the
new scenarios both improve the local climate and are comparable in terms of comfort, and reduce
average air temperatures by 0.5-1C. Scenario 2 provides a greater level of flow reduction for all
storms except 13/06/07, it also is the best option for reducing both average and peak local air
temperatures. This indicates that there is a clear link between SUDS and UHI alleviation, with the
inclusion of SUDS providing benefits for both. Furthermore, the scenario that provides the greatest
flow reduction also provides the greatest temperature reduction. However, the land take (area
required solely for the SUDS structures) required is much greater for this scenario. It must also be
noted, in terms of technical feasibility, that while Scenario 1 could theoretically be implemented in
incremental phases; Scenario 2 cannot due to the nature of the flood channel. Further work would
also require consideration of water quality benefits and consideration of the local climate in winter
conditions.


ACKNOWLEDGEMENTS
Many thanks to Laurence Pattacini for the Urban form designs.
This paper is based on work undertaken within the URSULA project, funded by the UK
Engineering and Physical Sciences Research Council (Grant number: P/F007388/1). The authors
are grateful for this support. The views presented in the paper are those of the authors and cannot be
taken as indicative in any way of the position of the funders or of colleagues and partners.

REFERENCES
Ali-Toudert, F; (2005) Dependence of Out Door Thermal Comfort on the Street Design in Hot and Dry Climate. Institute of
Meteorology: PhD. Thesis, Freiburg, Germany
Bastien, N; Arthur, S; Wallis, S; Scholz, M. (2010) The best management of SuDS treatment trains: a holistic approach. Wat Sci
and Tech, 61 (1), 263-271.
Berryman, J and Stovin, V.R (2010). Modelling a Green Roof and the benefits to Storm Water Management. World Green Roof
Congress, 15-16
th
September 2010, London.
Biesbroek, G.R., R.J. Swart, T.R. Carter, C. Cowan, T. Henrichs, H. Mela, M.D. Morecroft, and D. Rey (2010), Europe adapts to
climate change: Comparing National Adaptation Strategies. Global Environmental Change, 20(3): p. 440-450.
Bruse, M. (1999): Simulating microscale climate interactions in complex terrain with a high-resolution numerical model: A case
study for the Sydney CBD Area. In: Proceedings International Conference on Urban Climatology & International Congress of
Biometeorology, Sydney, 8-12. Nov, AustraliaCheng, M-S; Zhen, J.Z; Shoemaker, L (2009). BMP decision support system for
evaluating stormwater management alternatives. Front. Environ. Sci. Engin. China, 3 (4), 453-463.
Ellis, J.B; Deutsch, J.-C; Legret, M; Martin, C; Revitt, D.M; Scholes, L; Seiker, H and Zimmerman, U (2005). The DayWater
decision support approach to the selection of sustainable drainage systems: A mulit-criteria methodology for BMP decision
makers. . 10
th
ICUD, Copenhagen/Denmark, 21-26 August 2005
Echols, S (2007) Artful Rainwater Design. J. of Green Buildings, 2 (4), 103-122.
Fahmy, M. (2010). Interactive Urban Form Design of Local Climate Scale in Hot Semi-Arid Zone. Architecture. Sheffield,
University of Sheffeild. PhD Thesis: 254
Gober, P; Brazerl, A; Quay, R; Myint, S; Grossman-Clarke, S; Miller, A; Rossi, S. (2010) Using watered landscapes to manipulate
urban heat island effects: How much water will it take to cool Pheonix. Journal of the American Planning Association. 76 (1),
109-121.
Guthanker, S; Gober, P. (2007) The impact of Pheonix urban heat island on residential water use. Journal of the American Planning
Association, B73 (3), 317-329
Grimmond, C.S.B; Oke, T.R. (1999) Heat storage in urban areas: Local scale observations and evaluation of a simple mode. Journal
of Applied Meteorology. 38, 922-940
Howard, L. (1833) The Climate of London. Harvey and Darton, London
Jauregui, E., Influence of a large urban park on temperature and convective precipitation in a tropical city. Energy and Buildings,
1990-1991. 15-16: p. 457-463
Jendritzky, G., A. Maarouf, D. Fiala and H. Staiger (2002). "An Update on the Development of a Universal Thermal Climate Index."
15th Conf. Biomet. Aerobiol. and 16th ICB02, Kansas City 27 Oct - 1 Nov.
Natural England (2009) Green Infrastructure Guidance. Available from www.naturalengland.org.
Scholes, l; Ellis, J.B; Revitt, D.M (2005). The fate of stormwater priority pollutants in BMPs. Deliverable 5.3. Daywater.
Available from www.daywater.org.
Shashua-Bar, L. and M.E. Hoffman, Vegetation as a climatic component in the design of an urban street: An empirical model for
predicting the cooling effect of urban green areas with trees. Energy and Buildings, 2000. 31(3): p. 221-235.
Stahre, P (2008) Blue Green Fingerprints in the City of Malmo. VASyd Sweden.
Smith, C. Levermore, G. (2008) Designing urban spaces and buildings to improve sustainability and quality of life in a warmer
world. Energy Policy. 36(12), 4558-4562.
Stovin, V.R., and Swan, A.D. (2007). Retrofit SUDS cost estimates and decision support tools. Water Management, 160 207-214.
Upmanis, H., Eliasson, I., Lindqvist, S. (1998) The influence of green areas on nocturnal temperatures in a High Latitude City
(Goteborg, Sweden). International Journal of Climatology. 18: 681-700
Vandentorren, S., Bretin, P., Zeghnoun., et al. (2006) August 2003 Heat Wave in France: Risk Factors for Death of Elderly People
Living at Home. Eur J Public Health 16:6 583-591
Woods-Ballard, B; Kellagher, R; Martin, P; Jefferies, C; Bray, R; Shaffer, P (2007) The SUDS manual. CIRIA C697. London.