Hydrological Design and Analysis Vol 1

Eco-Hytech Research Centre 2021

ISBN xxxxxxxxxxxxxxxxxxxxxx

CHAPTER x

Evapotranspiration performance of green roof system using different types of

local vegetation

Hartini Kasmin*, Sitinurr Hidayu Baharuddin, Ahmad Afdhal Abd Aziz and Nur Durrah Sharif

Faculty of Civil and Environmental Engineering,

Universiti Tun Hussein Onn Malaysia, 86400, Batu Pahat, Malaysia

*Corresponding author email: [email protected]

Abstract

Green roof is a technology of incorporation plants with build environment. Green roof or
vegetated roof is a roof which is partially or completely covered with vegetation and soil or
growing medium. Evapotranspiration (ET) is one of a key parameter that influences the
stormwater retention capacity, and thus the hydrological performance of green roofs. This
paper investigates on how the moisture content in extensive green roofs varies during dry
periods due to evapotranspiration performance for a proposed green roof system using
different types of vegetation. The ET experiments were done in 2014, 2015 and 2016,
separately using three types of vegetation; Wedelia Tilobata, Portulaca sp, and Arachis Pintoi.
However, due to limited supply, Arachis Pintoi was substituted into Japanese Grass in 2016.
In each experiment, four different test beds were used; three test beds for three vegetation
and the other one was used as a control with substrate layer only. The weight of test beds
were measured and recorded daily. Changed in weight mass represents the water loss by ET
for vegetated roofs and evaporation (E) for the non-vegetated roof. It was observed that from
three series of ET experiments; after 5 days, Wedilia Tilobata test bed has shown better water
reduction than Portulaca sp. However, the percentage of reduction quite varies between
experiment in 2014 and 2015 although the temperature quite similar between 28˚C and 32˚C.
This may due to the system of drainage constructed for the green roof setup, the density of
the root and vegetation growth and the influence of other climatic condition such as wind and
radiation of the surrounding. The performance of Arachis Pintoi also quite varies between
these experiments, however, the performance of water reduction for Arachis Pintoi seems
better than Portulaca sp within the similar temperature; but the water losses of Japanese
Grass test bed in 2016 displayed the lowest between these other three vegetation. Still, when
compared to the non-vegetated roof, all four vegetation exhibited higher water loss, but
Japanese Grass test bed start to lose more than the non-vegetated after ten days of dry days,
or else it seems simulating the non-vegetated performance. It can be concluded that the
understanding on how the rate of evapotranspiration will differ between different types of
drought tolerance vegetation to local weather, would greatly benefit in developing a more
effective green roof structure.

Keywords—green roof; evaporation; evapotranspiration; urban stormwater

1. Introduction
Green roofs have several positive effects in the urban setting with the most important ones
being their ability to retain and detain storm water (Villarreal & Bengtsson, 2005), reduce urban heat
islands (Wong et al., 2003), reduce building energy consumption by cooling roofs during summer
months (Barrio & Palomo, 1998), and create habitats for certain plants and animals and thereby
improve urban biodiversity (Emilsson et al., 2007).

1
Hydrological Design and Analysis Vol 1
Eco-Hytech Research Centre 2021
ISBN xxxxxxxxxxxxxxxxxxxxxx
Green roof is defined as a roof that supports vegetation (Werthmann, 2007). It is also known
as eco roof, living roof, planted roof or vegetated roof which uses plants to improve roof performance
and appearance (Snodgrass et al., 2010). Green roof is also referred to as roof garden in some
places (Dunnett & Kingsbury, 2004). This kind of roof can be divided into two types which are
intensive and extensive green roofs. An intensive green roof requires deeper substrates which is at
least 15 cm (6 inches) deep and generally known as a roof garden while extensive green roofs require
shallow substrates of between 2 cm to 15 cm (0.8 to 6 inches). Intensive green roofs are usually
accessible to the public and require regular maintenance while extensive green roofs are normally
inaccessible to the public and require minimal maintenance (Dunnett & Kingsbury, 2004).
Green roofs could give many environmental benefits to the buildings and occupants including
improving thermal performance of buildings by providing large surfaces with vegetation
(Eomorfopoulou & Aravantinos, 1998). In 2001, (Niachou et al., 2001) published the results of a study
on indoor and outdoor microclimate before and after the installation of a green roof on a building
which concluded that the indoor temperature values in the building with green roof are lower during
the day. The results show that the surface temperature of a non-insulated building without green roof
vary from 42 ℃ to 48 ℃ while the surface temperature of a green roof on non-insulated building are
lower and ranging from 28 ℃ to 40 ℃ . The incidence of the large temperature differences was due
to the installation of a green roof and this could contribute to energy saving potential.
The amount of water that is stored in a green roof, and then evapotranspired into the
atmosphere, is dependent on the depth and type of growing medium, type of drainage layer,
vegetation used and regional weather but this study will focus on the types of vegetation only. Despite
constraints such as high capital and maintenance costs, green roofs have received increased
attention in view of their beneficial impacts. From the storm water management aspect, green roofs
can play an important role in modern urban drainage because of their ability to slow down and reduce
runoff volume. High evapotranspiration from a green roof can reduce the annual runoff to less than
half the precipitation (Liesecke, 1998; Knoll, 2000; Bengtsson et al., 2005).

1.1 Hydrological Cycle in Natural Catchment

The hydrologic cycle – the cycling of water among land, ocean, and air – is an intimate part of
weather, climate, and life (Figure 1.1). Water can be stored in any one of the following major
reservoirs: atmosphere, oceans, lakes, rivers, soils, glaciers, snowfields, and groundwater. Water
moves from one reservoir to another by way of processes like evaporation, condensation,
precipitation, deposition, runoff, infiltration, sublimation, transpiration, melting and groundwater flow.
The oceans supply most of the evaporated water found in the atmosphere. Of this evaporated water,
only 91% of it is returned to the ocean basins by way of precipitation. The remaining 9% is transported
to areas over landmasses where climatological factors induce the formation of precipitation. The
resulting imbalance between rates of evaporation and precipitation over land and ocean is corrected
by runoff and groundwater flow to the oceans (Hubbart, 2010).
Evaporation is the physical process by which liquid water in the oceans or on land is changed
to vapor in the air. It occurs when unsaturated air comes into contact with a moist surface.
Evaporation provides the atmospheric moisture that eventually returns to the surface as rain or snow.
Evaporation also consumes an enormous amount of heat, which helps to cool the evaporating
surface. Once in the atmosphere, water condenses, forming clouds, and if conditions are right the
water falls back to the surface as precipitation. As water vapor in the atmosphere condenses and
changes from vapor to liquid, heat is released. This heat is an important source of energy to drive
atmospheric circulation and fuel storms (Bonan, 2002).

2
Hydrological Design and Analysis Vol 1
Eco-Hytech Research Centre 2021
ISBN xxxxxxxxxxxxxxxxxxxxxx

Figure 1.1: Natural Hydrological Cycle

Transpiration is the process by which water moves through plants and evaporates through leaf
stomata which are small openings in the leaves. It is the process by which soil moisture taken up by
vegetation is eventually evaporated as it exists at plant pores. Transpiration rate is also influenced by
crop characteristics, environmental aspects and cultivation practices. Different kinds of plants may
have different transpiration rates. Not only the type of crop, but also the crop development,
environment and management should be considered when assessing transpiration.

Figure 1.2: Evaporation and transpiration process

(Source: Howard, 2014)

Evapotranspiration (ET) is a term used to describe the sum of evaporation and plant transpiration
from the earth’s land surface to atmosphere. Evaporation occurs when unsaturated air comes in
contact with a wet surface (e.g., lakes, rivers, wet soil). Transpiration is evaporation of water from
plant leaves as it moves from the soil through plants and out through their leaves to the air. Plants
consume large amounts of water during growth. Meteorological processes near the ground control
evaporation and transpiration. Transpiration is also regulated by the physiology of plants. When plants
cover a small portion of the surface, evaporation is the dominant flux. Transpiration becomes more
important as plant cover increases. However, it is difficult to distinguish evaporation from transpiration,
and the two terms are often combined into evapotranspiration. Water lost from leaves during
transpiration must be replenished from the soil (Bonan, 2002).

1.2 Hydrological Cycle in Urban Catchment

The hydrological characteristics of urban water systems are predominantly influenced
by a relatively large proportion of impervious surface area and the presence of man-made or
hydraulically improved drainage systems (Bedient et al., 2008). Because less pervious area is
available for infiltration and because unlined natural drainage channels are being replaced by lined

3
Hydrological Design and Analysis Vol 1
Eco-Hytech Research Centre 2021
ISBN xxxxxxxxxxxxxxxxxxxxxx
channels or storm sewers (Wong & Chen, 1989), the response of an urban catchment to rainfall is
much faster than the response of a rural area with comparable geophysical characteristics such as
area and slope (Bedient, 2008). Next to this decrease in lag time between rainfall peak and discharge
peak, urban development also increases the runoff volume and runoff peak. The combined effects of
urbanization on the rainfall-runoff increases the exposure to the hazard of flooding (Mansell, 2003)
and erosion.
Urbanization produces numerous changes in the natural environment. It is predicted to rise
from 75% of people in developed countries in 2000 to 83% in 2030, while over the same period it will
rise from 40% to 56% in less developed countries (Cohen, 2003). Numerous factors influence the
amount of runoff but an indicative finding from (Rose & Peters, 2001) is that peak flows are from 30%
to more than 100% greater in urbanized catchments compared to the less urbanized and non-
urbanized catchments. In addition, in accordance with Manning’s equation which indicates that the
velocity of flow of water is indirectly proportional to the roughness of the land surface (Leopold et al.,
1995), stormwater flows more rapidly across smooth urban surfaces than across rough natural
surfaces.
When raining in urban area, the rainwater cannot seek into the ground and cause immediate
runoff in a high velocity because the imperviousness in urban system varies with lot size of residential
areas (Leopold, 1968), like the open surface like parking lots, roof tops and pavement are
impermeable. Therefore, water will not take time to fill up the streams. High discharge occur rapidly
and the lag time of the stream also will be shorted. This causes urban streams and catchments to be
more "flashy", where large flow events occur more frequently with faster ascending and descending
rates in the hydrograph (Walsh et al., 2005). Therefore, urban areas are more prone to flooding
compare to the rural area.
Figure 1.3 shows the difference between hydrological cycle in a natural and urban catchment
area. The percentage values indicate qualitative differences in the size of annual water volumes
through each pathway. Water that falls on the catchment area and is not evaporate or transpired may
reach the stream by three possible paths: overland flow/ runoff (almost all of which is transmitted to
the stream by storm water pipes in the urban catchment), subsurface flow through permeable topsoil/
shallow infiltration, or percolation into groundwater flow/ deep infiltration (Walsh et al., 2005).

Figure 1.3: The percentage of hydrological cycle in a natural catchment and in an urbanized
catchment (MMSD, 2016)

4
Hydrological Design and Analysis Vol 1
Eco-Hytech Research Centre 2021
ISBN xxxxxxxxxxxxxxxxxxxxxx
Urban catchments have a complicated water cycle involving water supply, wastewater
disposal and stormwater drainage. High quality water is harvested from catchments often some
distance away from urban areas, treated and delivered to meet domestic and industrial water
demands. Wastewater generated from urban areas is conveyed to regional wastewater treatment
facilities and then discharged to the environment (for example, to rivers or bays). Stormwater
generated from urban areas is often conveyed efficiently to designate trunk stormwater drainage
systems to reduce stormwater ponding and flooding (Lloyd et al., 2002).

1.3 Sustainable Urban Drainage System (SUDS)

Traditional storm water management in urban areas aims at removing runoff as quickly as
possible, eventually gathering excess runoff in detention basins for peak reduction. The conventional
approach to storm water management involves the efficient capture, convey and sometimes treat
runoff generated from the impervious surfaces. Sustainable Urban Drainage Systems (SUDS), Low
Impact Development (LID) or Water Sensitive Urban Design (WSUD) principles and applications are a
relatively new approach to storm water management that represent a conceptual improvement from
the conventional sewer system design methodology (Palla et al., 2010).
These principle are basically a source in reduction approach. These sustainable strategies
focus on evaporating, transpiring and infiltrating storm water on-site via natural or engineered soil,
vegetation and bioengineering application in order to reduce and treat the surface runoff that is limited
for most rainfall event in a natural environment.
The SUDS is systems that use the natural approach to manage the drainage system by
slowing and holding back the runoff water that come from site and allows the natural process to
breakdown pollutant. This system is developed are to control the problem of surface runoff that
caused the flash flood in urban area because of this area do not have natural element such as trees
to control the rate of water runoff.
SUDS consists of natural-storage-based structures such as swale, detention basins and the
green roof. These elements use the natural approach by slowing and holding back the water runoff.
The benefits of this system are reduces flood risk, maintaining or restoring natural flow cycle,
minimizing diffuse pollution and improving water resources. This system effectively reduces flood risk
by control the water runoff flow. The water is slowing and holding back by natural element such as
trees and vegetation.

1.4 Green roof system

As a SUDS device, the function of green roofs is to reduce storm runoff at source. To have
permeable green plants within impermeable urban areas might offer some benefits towards restoring
a natural water cycle. The configurations of a green roof, which simply replicate the actual function of
a hydrological mechanism include the following (Figure 1.4):
1. The slope
2. A vegetation/plant layer
3. A substrate layer and the depth
4. A drainage layer

Green roofs (vegetated roof surfaces) are increasingly seen in cities because they are an
important strategy that addresses some key urban environmental issues. Green roofs can reduce
surface water runoff, provide a habitat for wildlife, moderate the urban heat island effect, improve
building insulation and energy efficiency, improve the air quality, create aesthetic and amenity value,
provide opportunities for urban food production and preserve the roof’s waterproofing (English Nature,
2003; Dunnett and Kingsbury, 2004).
Urban environments are characterized by impermeable surfaces which result in flash floods,
erosion and pollutant loading to surface waters (Jennings and Jarnagin, 2002). The substitution of
vegetated roofs for conventional roofs has the potential to reduce roof runoff by 100% in some cases
(Oberndorfer et al., 2007), with regional reductions of up to 2.7% predicted in a scenario of 10% green
roof coverage (Mentens et al., 2006). Some studies show that the reduction in runoff from green roofs
is largely attributable to the thickness of the substrate or slope of the roof as opposed to any direct
effect of the vegetation (Van Woert et al., 2005; Getter et al., 2007), but plant uptake and transpiration

5
Hydrological Design and Analysis Vol 1
Eco-Hytech Research Centre 2021
ISBN xxxxxxxxxxxxxxxxxxxxxx
should increase the ability of a roof to hold water from repeated rain events (Villarreal & Bengtsson,
2005). Maximizing transpiration from the roof may thus increase the total amount of water prevented
from running off into the sewer system.

2. Materials and Methods

In this study, four small scale roofs known as test beds were used in this experiment. The test
beds have the same rectangular size and depth of 13 cm (depth) x 31cm (length) x 24.5cm (width) of
plastic containers. The weight ratio for soil media was 1:1:1 where 1 kg of vermiculate soil, 1 kg of
perlite soil and 1 kg of peat moss soil were used. The test roofs were then filled in with the soil mixture
about 10 cm depth of soil content.
One test bed was filled with soil only (non-vegetated). Another three test beds were planted
with drought tolerance species vegetation which were Arachis Pintoi/ Zoysia Japonica, Wedilia
Trilobata and Protulaca sp. This types of vegetation are suitable to be growth in Malaysia climatic
conditiion. The roof with soil layer only (non-vegetated) would act as control green roof, to observe the
differentiation of the absence plant towards the quantity of moisture.
There were three layers of green roof constructiont. The upper layer was a layer for selected
vegetation for each roofs with different vegetation. Firstly the Arachis Pintoi that was substituted by
Zoysia Japonica later in 2016, followed by the Wedilia Trilobata and Protulaca sp. Plants were grown
to a minimum of 60% plant cover. For the second layer, there was substrate layer which the mixture of
vermiculite, perlite and peat moss soil. The bottom layer was drainage layer using geotextile.
Geotextile provide an excellent drainage system, where it allows the flow of water through the soil
without causing the clogging or making soil become unsaturated.

(a) (b)

Figure 2.1: Diagram of (a) non-vegetated roof bed; (b) vegetated roof bed with Arachis Pintoi;

2.1 Evapotranspiration Test

On the day the tests began, the test beds were saturated for 24 hours then drained for 2 hours
to ensure substrates were at maximum field capacity. After 2 hours of water drained, the weight of test
beds were then measured and recorded daily. Change in mass weight represents water loss by ET for
vegetated roofs and by E for bare substrate. Transpiration, T was determined for each vegetation as
the weight difference between measured ET and E. All test beds were placed under shaded roof but
in open area (to avoid another moisture from re-filled in the test beds but still having influence from
the surrounding climatic parameters) for natural ET and E process. Temperatures were recorded
daily.
The ET tests in 2015 and 2016 were then modified from a similar study by Maclvor &
Lundholm (2011). The objective of the test was to identify water loss rate from different types of
vegetation. Vegetation with the highest evapotranspiration rate would likely retain more water during a
rainfall period. Water loss is determined by weighing green roof modules in an interval of hours with
digital bench scale. Difference in weight is considered as water loss due to evapotranspiration
(Krishnan et al., 2014).
Several methods are available to estimate evapotranspiration. One simple approach is to
monitor the water level in large shallow pans. In the absence of rainfall, the change in water level is
the evaporative loss. Another method is based on the water balance. Consider, for example, a volume
of soil with water input from precipitation and water loss from drainage and evapotranspiration. The
water balance is ∆ S = P − R – ET. Evapotranspiration can be obtained from direct observations as
the residual if all other terms in the water balance are known (Bonan, 2002).

2.2 Vegetation Layers

6
Hydrological Design and Analysis Vol 1
Eco-Hytech Research Centre 2021
ISBN xxxxxxxxxxxxxxxxxxxxxx
Difference in water uptake has direct relationship towards differences in stormwater runoff
reduction. Therefore, significantly affecting the ability for the substrate to absorb and retain more
water during a rainfall event. An experiment done by (Berghahe, 2007) showed evapotranspiration
rate difference between plants characteristic could be up to 40% (Krishnan et al., 2014).
Tropical climate country like Malaysia would definitely benefited from this study as stormwater
event runoff is a common phenomenon that has been continuously causing mass destruction recently.
For this study, 4 types of drought tolerance vegetation namely Wedelia Trilobata, Portulaca Sp,
Arachis Pintoi and Japanese grass were selected. These four species were selected based on their
high survival characteristic which can cope with the tropical climate of Malaysia and be easily
maintained.

2.2.1 Wedelia Trilobata

It would be hard to find another groundcover better suited to hot, dry conditions than Wedelia
(Figure 2.4). Attractive, glossy, dark green, lobed leaves, rapidly spreading growth habit, and a
continuous display of small, bright yellow, daisy-like blooms create a much-favored landscape plant.
Suited to a wide variety of conditions, Wedelia will cover rough, rocky ground or wet drainage ditches,
and even tolerates some degree of foot traffic. Wedelia has a vine-like habit and will grow up into
shrubs and trees planted in the bed. Like ivy and other creepers, it will require regular trimming along
the edge of the groundcover bed to control its spread (Gilman, 2014).

Figure 2.1: Wedelia Trilobata

2.2.2 Portulaca Sp
Also known as Moss Rose, Rose Moss, Portulaca grandiflora. It comprises of about 40-100
species and commonly found in the tropics and warm temperate regions. The botanical name is
derived from the Latin Potare, meaning to “carry,” and Lac or “milk”, referring to the milky sap of the
plant (Hocking, 1997). Portulaca (Portulaca spp.) is a non-fussy plant and valued for its attractive
foliage and brilliant flowers. They have been used as a folk medicine in many countries for skin
rashes, inflammation, ulcers, abdominal complaints, detoxification, cough and urinary discharge
(Krittikar et al., 2001, Grieve 1999, Grubben, 2004).

Figure 3.2: Portulaca sp.

7
Hydrological Design and Analysis Vol 1
Eco-Hytech Research Centre 2021
ISBN xxxxxxxxxxxxxxxxxxxxxx

Figure 4.5: Japanese Grass

3. Results and Discussion

Figure 3.1 shows the results of daily soil moisture deficits for the five type of treatment in
different year of 2014, 2015 and 2016 with different average temperatures at 29.9˚C, 28.5˚C and
33.4˚C, respectively. Soil moisture deficit represents the evapotranspiration process without any
additional moisture during experiment. It was observed that temperature parameter alone does not
directly affected the soil moisture deficits. This figure highlighted the performance of all treatments
after five days where both in 2015 and 2016 shows similar results which have moisture reduction less
than 25 mm, whereas in 2014, the reduction of moistures after 5 days within the four test bed roofs
seems very high between 39 and 55 mm. The high difference of moisture reduction between 2014
and the other two years may be due to the method used during each experiment. In 2014, to simulate
the maximum water capacity holding for the soil mixture, the test bed roofs were completely
submerged in water for 24 hours, drained for 2 hours before any measurement started. While for the
experiments in 2015 and 2016, the method has changed due to lack of containers for submerging the
roofs within the same time; by filling the water into the test bed roofs until it started to generate runoff
and stop; and the observation and weight measurement started. Therefore, the level of saturation may
be different.
Looking at the performance for each vegetation, in 2015 and 2016, the transpiration process
seems to occur in all vegetation beds as predicted, except in 2014. In 2014, Wedelia Trilobata seems
to perform the ET process less than evaporation in non-vegetated roof bed. This may be due to the
foliage growth of Wedilia Trilobata was very less than other vegetated roofs. However due to the
saturation method used for maximum capacity holding in 2014, may be also contributed on higher
infiltration of moisture within few hours after the test beds were drained off.
However, the percentage of reduction quite varies between experiment in 2014 and 2015
although the averages temperature quite similar between 29.9˚C and 28.5˚C. This may due to the
system of drainage constructed for the green roof setup, the density of the root and vegetation growth
and the influence of other climatic condition such as wind and radiation of the surrounding. The
performance of Arachis Pintoi also quite varies between these experiments, however, the
performance of water reduction for Arachis Pintoi seems better than Portulaca sp within the similar
temperature; but the water losses of Japanese Grass test bed in 2016 displayed the lowest between
these other three vegetation. Still, when compared to the non-vegetated roof, all four vegetation
exhibited higher water loss, but Japanese Grass test bed start to lose more than the non-vegetated
after ten days of dry days, or else it seems simulating the non-vegetated performance.
Due to the non-uniform foliage growth for each vegetation in the three series of ET
experiments, therefore the performance observed for each vegetation were not similar. Therefore no
specific vegetation is better than the other one, except the fertile the vegetation growth may contribute
better ET than the other. In future, it is suggested that for full comparison can be made between
vegetation is only when the plant growth of each vegetation is in similar and in adequate amounts and
proportion.

8
Hydrological Design and Analysis Vol 1
Eco-Hytech Research Centre 2021
ISBN xxxxxxxxxxxxxxxxxxxxxx

0 1 2 3 4 5 6 0 1 2 3 4 5
0 0
Non-
vegetated
-10 Portulaca -10
Soil moisture deficits (mm)

Soil moisture deficits (mm)

Sp
-20 Wedilia -20
Trilobata

-30 -30

Non-vegetated
-40 -40
Portulaca Sp
-50 -50 Wedilia Trilobata

Arachis Pintoi
-60 -60
Time (days) Time (days)
(a) (b)

Figure 3.1: The soil moisture deficits of five roof treatment types (a) for 6 days in 2014; (b) for 5 days
in 2015;

4. Conclusions
This study has demonstrated that Wedelia Trilobata, Arachis Pintoi, and Japanese Grass are the
drought-resistant vegetation types and had shown that they could retained stormwater. After storing
water until it reached maximum capacity holding for the soil mixture, the water will loss through
evapotranspiration for vegetated soil and through evaporation for the non-vegetated soil depending on
the climatic condition of the surrounding area. Based on the performance of vegetated roof bed, the
four vegetation had significantly performed the ET process.

There are several recommendation can be proposed to improve the study:

1. The ET experiment can be started after all vegetation have fully growth in adequate and uniform
amounts.
2. Additional meteorological parameter such as relative humidity and wind should be measured and
observed for more accuracy ET data.
3. Different soil mixture can be used in future research in order to compare the results. Soil mixture
must able to contain high WCmax and high nutrient to provide for the vegetation. In this study, the
vegetation produces less flowers therefore, different soil mixture may be able to provide different
results.
4. Different method of analyzing moisture index also can be used in determining the Potential ET
rates which is by using Thornthwaite equation; Penman equation.

It can be concluded that the understanding on how the rate of evapotranspiration will differ between
different types of drought tolerance vegetation to local weather, would greatly benefit in developing a
more effective green roof structure.

References [1] Barrio, D., & Palomo, E. (1998). Analysis of

the green roofs cooling potential in buildings.

9
Hydrological Design and Analysis Vol 1
Eco-Hytech Research Centre 2021
ISBN xxxxxxxxxxxxxxxxxxxxxx
Energy and Buildings, 179-193. mid-Atlantic subwatershed.  Landscape
[2] Bedient, P.B., C.H. Huber, en B.E. Vieux, Ecology, 17(5), 471-489.
(2008). Hydrology and floodplain analysis. [15] Knoll, S. (2000). The runoff of extensively
Upper Saddle River, NJ: Prentice Hall, Inc. greened roofs. Master Thesis, IWAWI
Mitteilung Nr. 136, Technical University of
[3] Bengtsson, L., Grahn, L., & Olsson, J. Berlin
(2005). Hydrological function of a thin
extensive green roof in southern [16] Leopold, L.B. (1968).  Hydrology for Urban
Sweden. Nordic Hydrology, 36, 259-268. Land Planning – A Guidebook on the
[4] Bonan, G. (2002). Chapter 5 Hydrologic Hydrologic Effects of Urban Land Use.
Cycle. In Ecological Climatology by Gordon Geological Survey Circular 554,
Bonan (pp. 1-33). Cambridge University Washington.
Press
[5] Cohen, J. (2003). Human Population: The [17] Leopold, L.B., Wolman, M.G., & Miller, J.P.
Next Half Century. Science, 1172-1175. (1995).  Fluvial processes in geomorphology.
[6] Dunnett N. & Kingsbury N. (2004). Planting Dover Earth Science. Dover Publications;
Green Roofs and Living Walls. Timber Trade Paperback Edition.
Press. ISBN: 0-88192-640-X 2004, pp. 2-4. [18] Liesecke, H.J., (1998). Das
[7] Emilsson, T., Berndtsson, J., Mattsson, J., & retentionsvermo¨gen von dachbegru¨
Rolf, K. (2007). Effect of using conventional nungen (Water retention capacity of
and controlled release fertiliser on nutrient vegetated roofs). Stadt und Gru¨n 47,
runoff from various vegetated roof systems. 46e53.
Ecological Engineering, 260-271.
[8] English Nature (2003). Green roofs: Their [19] Lloyd, S.D., Wong, T.H., & Chesterfield,
existing status and potential for conserving C.J. (2002). Water sensitive urban design: a
biodiversity in urban areas. Peterborough: stormwater management perspective.
English Nature. [20] Maclvor J.S. & Lundholm J. (2011).
[9] Eomorfopoulou. E & Aravantinos D. (1998). Performance evaluation of native plants
“The Contribution of a Planted Roof to the suited to extensive green roof conditions in a
Thermal Protection of Buildings in Greece,” maritime climate. Ecological Engineering 37,
Energy and Buildings, Vol. 27, Feb.1998, pp 407-417
pp. 29-36.
[10] FAO, Food and Agriculture Organisation
[21] Mansell, M.G.(2003) Rural and urban
(2015). Arachis Pintoi Krap. & Greg.,
hydrology. London: Thomas Telford
Retrieved at
Publishing. Marsalek,
http://www.fao.org/ag/agp/agpc/doc/gbase/D
ATA/pf000463.htm
[22] Mentens, J., Raes, D., & Hermy, M. (2006).
Green roofs as a tool for solving the
[11] Getter, K. L., Rowe, D. B., & Andresen, J. A.
rainwater runoff problem in the urbanized
(2007). Quantifying the effect of slope on
21st century? Landscape and urban
extensive green roof stormwater
planning, 77(3), 217-226.
retention. Ecological Engineering, 31(4), 225
- 231. [23] MMSD, Milwaukee Metropolitan Sewerage
District (2016). Stormwater Trees. Retrieved
from https://www.mmsd.com/what-we-
[12] Howard P., (2014). Evapotranspiration, do/green-infrastructure/stormwater-trees or
Evapotranspiration - The Water Cycle. http://jh402is2104.blogspot.my/
Retrieved October 20, 2014 from
https://water.usgs.gov/edu/watercycleevapot
ranspiration.html [24] Niachou A., Papakonstantinou K.,
Santamouris M., Tsangrassoulis A. and
Mihalakakou G. (2001). “Analysis of the
[13] Hubbart. J. A., (2010). Hydrological Cycle Green Roof Thermal Properties and
in Encyclopedia of Earth. Retrieve from Investigation of Its Energy Performance,”
http://www.eoearth.org/view/article/153627/ Energy and Buildings, Vol. 33, Sep. 2001,
pp.719-729.
[14] Jennings, D.B., & Jarnagin, S.T. (2002).
Changes in anthropogenic impervious
surfaces, precipitation and daily streamflow
discharge: a historical perspective in a

10