Reducing Urban Heat Islands: Green Roofs

8/9/2019 Reducing Urban Heat Islands: Green Roofs

1/29

Reducing Urban Heat IslandsCompendium of Strategies

Green Roofs


2/29

Acknowledgements

Reducing Urban Heat Islands: Compendium of Strategiesdescribes the

causes and impacts o summertime urban heat islands and promotes

strategies or lowering temperatures in U.S. communities. This compendium

was developed by the Climate Protection Partnership Division in the U.S.

Environmental Protection Agencys Oce o Atmospheric Programs. Eva

Wong managed its overall development. Kathleen Hogan, Julie Rosenberg,

and Andrea Denny provided editorial support. Numerous EPA sta in

oces throughout the Agency contributed content and provided reviews.

Subject area experts rom other organizations around the United States and

Canada also committed their time to provide technical eedback.

Under contracts 68-W-02-029 and EP-C-06-003, Perrin Quarles Associates,

Inc. provided technical and administrative support or the entire

compendium, and Eastern Research Group, Inc. provided graphics and

production services.

PositvEnergy provided support in preparing the Trees and Vegetation, Cool

Roos, and UHI Activities chapters under contract PO #2W-0361-SATX.

Experts who helped shape this chapter include:

Ryan Bell, Robert Berghage, Hitesh Doshi, Robert Goo, David Hitchcock,

Megan Lewis, Tom Liptan, Karen Liu, Greg McPherson, Dave Nowak, Steven

Peck, Katrin Scholz-Barth, Je Sonne, Benjamin Taube, Linda Velazquez,

Kathy Wol, Jim Yarbrough, and Barry Zalph.


3/29

Contents

Green Roos 1

1. How It Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2

2. Green Roo Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2.1 Extensive Green Roos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4

2.2 Intensive Green Roos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4

3. Benets and Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3.1 Benets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3.2 Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10

3.3 Benet-Cost Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11

4. Other Factors to Consider . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13

4.1 Site Characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13

4.2 Installation and Maintenance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14

( L o w - P r o l e / E c o r o o s ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17

(High-Prole/Roo Gardens) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17

4.3 Fire Saety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

5. Green Roo Initiatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

6. Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22

Endnotes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23


4/29

Green Roos

Green roos are an emerging technol

ogy that can help communities miti

gate urban heat islands. A green roo

is a vegetative layer grown on a rootop.

As with trees and vegetation elsewhere,

vegetation on a green roo shades suraces

and removes heat rom the air through

evapotranspiration. These two mechanisms

reduce temperatures o the roo suraceand the surrounding air. The surace o a

vegetated rootop can be cooler than the

ambient air, whereas conventional rootop

suraces can exceed ambient air tempera

tures by up to 90F (50C).2 Green roos

can be installed on a wide range o build

ings, including industrial, educational,

and government acilities; oces; other

commercial property; and residences. This

chapter reviews:

How green roos work to mitigateheat islands

What types o green roos are available The benets and costs o green roos Other actors to consider in using this

mitigation strategy

Initiatives used to promote green roos Tools and resources to urther explore

this technology.

Opportunities to Expand Use oGreen Roos in Urban Areas

Most U.S. cities have signicant opportunities to

increase the use o green roos. As part o EPAs

Urban Heat Island Pilot Project, the Lawrence

Berkeley National Laboratory conducted analysesto estimate baseline land use and tree cover in

ormation or the pilot program cities.1 Figure 1

shows the percentage o roo cover in our o these

urban areas: roos account or 20 to 25 percent o

land cover. Even though not all these areas will be

likely candidates or installing a green roo, there

is a large opportunity to use green roos or heat

island mitigation.

0 5 10 15 20 25

Chicago

Houston

Sacramento

Salt Lake City

Percent Coverage

Figure 1: Roo Cover Statistics or Four U.S. Cities

(Below Tree Canopy)

3

GREEN ROOFS DRAFT 1


5/29

1 How It Works

With regard to urban heat islands, green

roos work by shading roo suraces and

through evapotranspiration. Using green

roos throughout a city can help reduce

surace urban heat islands and cool the air.

Shading.The plants o a green roo and

the associated growing medium, a specially

engineered soil, block sunlight rom reach

ing the underlying roo membrane. Though

trees and vines may not be common on

green roos, they indicate how other vege

tation on green roos shade suraces below

them. For example, the amount o sunlight

transmitted through the canopy o a tree

will vary by species. In the summertime,generally only 10 to 30 percent o the suns

energy reaches the area below a tree, with

Green Roo Market

In the United States demand and

interest in green roos has grown

tremendously. A survey o Green

Roos or Healthy Cities membersound that 25 percent more square

eet o green roong were installed

in the United States in 2005 than in

2004.3A Green Roos Project Data

base, available at , estimated

a total o 6.6 million square eet

(614,000 m2) o completed or ongo

ing green roo projects in the United

States as o June 2007. Germany,

widely considered a leader in green

roo research, technology, and usage,

has had decades o experience with

green roos. An estimated 10 percent

o all fat roos in Germany are roo

top gardens.4,5

Figure 2: Intensive Green Roo in

Frankurt, Germany

Germany has long been a leader in green roos; an

intensive green roo covers much o this building

in Frankurt.

WikimediaCommons

the remainder being absorbed by leavesand used or photosynthesis and some

being refected back into the atmosphere.

In winter, the range o sunlight transmit

ted through a tree is much wider10 to 80

percentbecause evergreen and decidu

ous trees have dierent wintertime oliage,

with deciduous trees losing the leaves and

allowing more sunlight through.6

Shading reduces surace temperatures

below the plants. These cooler suraces,in turn, reduce the heat transmitted into

buildings or re-emitted into the atmo

sphere. For example, a multi-month study

measured maximum surace temperature

reductions due to shade trees ranging rom

20 to 45F (11-25 C) or walls and roos

at two buildings.7Another study examined

the eects o vines on wall temperatures,

and ound reductions o up to 36F (20C).8

Furthermore, the growing medium o a

green roo itsel protects the underlyinglayers rom exposure to wind and ultravio

let radiation.

REDUCING URBAN HEAT ISLANDS DRAFT2
http://www.greenroofs.com/projects/plist.phphttp://www.greenroofs.com/projects/plist.phphttp://www.greenroofs.com/projects/plist.phphttp://www.greenroofs.com/projects/plist.php


6/29

Evapotranspiration. Plants absorb water

through their roots and emit it through their

leavesthis movement o water is called

transpiration. Evaporation, the conversion

o water rom a liquid to a gas, also occurs

rom the suraces o vegetation and the sur

rounding growing medium. Together, the

processes o evaporation and transpiration

are reerred to as evapotranspiration. Evapo

transpiration cools the air by using heat

rom the air to evaporate water.

Figure 3: Evapotranspiration and

Shading on a Green Roo

Evapotranspiration

Shading

Indoor

Thermal mass

KarenLiu

Plant shade reduces the sunlight that reaches the

roo. Evapotranspiration urther cools a green roo

by using heat to evaporate water rom the growing

medium and plant suraces

Green roo temperatures depend on the

roos composition, moisture content o the

growing medium, geographic location, so

lar exposure, and other site-specic actors.

Through shading and evapotranspiration,

most green roo suraces stay cooler than

conventional rootops under summertime

conditions. Numerous communities and

research centers have compared surace

temperatures between green and conventional roos. For example:

Chicago compared summertime suracetemperatures on a green roo with a

neighboring building. On an August

day in the early aternoon, with tem

peratures in the 90s, the green roo

surace temperature ranged rom 91 to

119F (33 to 48C), while the dark, con

ventional roo o the adjacent building

was 169F (76C). The near-surace air

temperature above the green roo was

about 7F (4C) cooler than that over

the conventional roo.9

A similar study in Florida ound thatthe average maximum surace tempera

ture o a green roo was 86F (30C)

while the adjacent light-colored roo

was 134F (57C).10

Reduced surace temperatures help build

ings stay cooler because less heat fows

through the roo and into the building. In

addition, lower green roo temperaturesresult in less heat transer to the air above

the roo, which can help keep urban air

temperatures lower as well. Some analyses

have attempted to quantiy the potential

temperature reductions over a broad area

rom widespread adoption o green roo

technology. A modeling study or Toronto,

Canada, or example, predicted that adding

green roos to 50 percent o the available

suraces downtown would cool the entire

city by 0.2 to 1.4F (0.1 to 0.8C). Irrigatingthese roos could urther reduce tempera

tures by about 3.5F (2C) and extend a 1

to 2F (0.5-1C) cooled area over a larger

geographic region. The simulation showed

that, especially with sucient moisture or

evaporative cooling, green roos could play

a role in reducing atmospheric urban heat

islands.11

A similar study in New York City modeled

air temperature reductions two meters, or6.5 eet, above the roo surace based on a

scenario assuming 100 percent conversion

o all available roos area to green roos.

The model results estimated a temperature

reduction o about 0.4F (0.2C) or the city

as a whole, averaged over all times o the

day. The model projected that temperatures

GREEN ROOFS DRAFT 3


7/29

Figure 4: Temperature Dierences between a Green and Conventional Roo

NationalCenterofExcellence/ASU

On a typical day, the Chicago City Hall green roo measures almost 80F (40C) cooler than the neighboring conventional roo.

at three oclock in the aternoon would

be reduced 0.8F (0.4C). The researchers

also evaluated, in detail, six areas withinthe city. The area with the highest 24-hour

average reduction in temperature had a

change o 1.1F (0.6C), and the reductions

at three oclock in the aternoon in those

six areas ranged rom 0.8F (0.4C) to 1.8F

(1.0C).12

2 Green Roo Types

A green roo can be as simple as a 2-inch

(5 cm) covering o hardy, alpine-likegroundcover, generally termed an exten

sive system, or as complex as a ully ac

cessible park complete with trees, called an

intensive system.

2.1 Extensive Green Roos

For the simpler, lighter weight extensive

green roo system, plant selections typi

cally include sedumssucculent, hardy

plantsand other vegetation generally suit

able or an alpine environment. The concept is to design a rugged green roo that

needs little maintenance or human interven

tion once it is established. Plants adapted to

extreme climates oten make good choices

and may not require permanent irrigation

systems. Overall, because o their light

weight, extensive systems will require the

least amount o added structural support,

which improves their cost-eectiveness

when retrotting an existing structure.

Extensive green roos have been grown on

roos with slopes o 30 or more, which

would equal a ratio o rise to run o 7:12 or

greater. (In contrast, a low-sloped roo with

a ratio o rise to run o 2:12 would have a

slope o 9.5.) The slope determines i the

roo will need additional support to hold

the growing medium and other parts o the

vegetative layer in place. Steeper roos may

retain less stormwater than an equivalent,fatter roo.

2.2 Intensive Green Roos

An intensive green roois like a conven

tional garden, or park, with almost no limit

on the type o available plants, including

large trees and shrubs. Building owners or

managers oten install these roos to save

energy and provide a garden environment

or the building occupants or the general

public to enjoy. Compared to extensive

green roos, intensive green roos are

heavier and require a higher initial invest

ment and more maintenance over the long

term than extensive roos. They generally

require more structural support to ac

commodate the weight o the additional

growing medium and public use. Intensive



8/29

Figure 5: Combination Extensive/ Figure 6: Fords Dearborn Truck Plant: An

Intensive Green RooThe Rootop Example o an Extensive Green Roo

Garden on Chicagos City Hall

J.DavidMattox/CityofManhattan,

Kansa

s

Fords Dearborn Truck Plant in Michigan covers 10.4

acres (42,100 m2) and is anticipated to reduce the

buildings energy costs by 7 percent.15

GreenRoofsforHealthyCities/ww

w.g

reenroofs

.org

The photograph provides an example o a

combination extensive/intensive green roo on

Chicagos City Hall.

systems also need to employ irrigationin the Trees and Vegetation chapter andare briefy described here in the context o

systems, which can use rainwater captured

rom the roo or another source.

3 Benefts and Costs

Green roos provide many o the same

benets that trees and other ground level

vegetation provide. Green roos have an

advantage, though, in that they can be

used in dense, built-up areas that may nothave space or planting at the ground level.

The benets o vegetation were discussed

Green Roos and Green

Walls

In addition to green roos, build

ing owners can install green walls,

sometimes reerred to as living wallsor vertical gardens. These walls can

involve placing trellises or cables in

ront o exterior walls and allowing

vines to grow up them, or can be

more elaborate, with plants actually

incorporated into the wall.13

green roos.

3.1 Benets

Reduced Energy Use. Green roos can

save energy needed to cool and heat

the buildings they shelter. When green

roos are wet, they absorb and store large

amounts o heat, which reduces tempera

ture fuctuations. When dry, green roo lay

ers act as an insulator, decreasing the fow

o heat through the roo, thereby reducing

the cooling energy needed to reduce build

ing interior temperatures. In the winter,

this insulating eect means that less heat

rom inside the building is lost through the

roo, which reduces heating needs. In the

summertime, green roo vegetation reduces

roo surace temperatures and ambient air

temperatures, thus lowering cooling energy

demand. The insulating properties o green

roos vary as they are dynamic systems that

change throughout the year, particularly

with regard to water storage. As with cool

roos, discussed in the Cool Roo chapter,

green roos should not be used as a substi

tute or insulation.

GREEN ROOFS DRAFT 5
http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/http://www.greenroofs.org/


9/29

Figure 7: Green Wall in Huntsville, Alabama

This 2,000-square oot (190 m2) green wall on a store

in Huntsville, Alabama, is one o the largest in North

America. 14

EOA/ElmslieOslerArchitect

Green Roo TypesChanging Nomenclature?

The term low prole has been used

in place o extensive to describe

green roos that are lighter weight,

shallower, and simpler. Similarly,

high prole or deep prole hasbeen used instead o intensive to

describe a heavier, more complex

green roo system with deeper soil.

Figure 8 compares the average daily fow

o heat through a dark, conventional roo

and an extensive green roo in Ottawa,

Canada. During the spring and summer,

rom May to September 2001, the energy

demand needed to remove heat that fowedthrough the conventional roo was six to

eight kilowatt hours (kWh) a day, while the

green roos energy demand rom heat fow

was less than 1.5 kWh a day, a reduction o

more than 75 percent. In contrast, during

the all and winter months, rom Novem

ber 2000 through March 2001, heat fow

through the green roo was only slightly

less than the reerence roo in all months

except January, so that the energy demand

rom both roos was relatively similar.

During this time, snow had accumulated,

and the temperatures o both roos stayed

about the same.16

Although green roos can save energy both

in summer and winter, the specic savings

will depend on the local climate and individ

ual building and roo characteristics, such as

size, use, and insulation. For example:

Chicago estimates that its City Hallgreen roo project could provide cool

ing savings o approximately 9,270

kWh per year and heating savings o

740 million Btus.18This translates into

annual, building-level energy savings o

about $3,600.

A Canadian study modeled the heatingand cooling energy savings o a roughly

32,000- square oot (2,980 m2) green

roo on a one-story commercial build

ing in Toronto.19The analysis estimated

that the green roo could save about 6

percent o total cooling and 10 percent

o heating energy usage, respectively,or about 21,000 kWh total. The study

noted that the cooling energy savings

would be greater in lower latitudes.

For instance, when the authors ran the

same simulation or Santa Barbara, Cali

ornia, the cooling savings increased to

10 percent.

A study in central Florida measuredyear-round energy savings rom a green

roo. By the roos second summer, the

average rate o heat transer, or fux,through the green roo was more than

40 percent less than or the adjacent

light-colored roo. The reduced heat

fux was roughly estimated to lower

summertime energy consumption o

the 3,300 square oot (1,000 m2) project

building by approximately 2.0 kWh per

day.20 Under winter heating conditions,



10/29

Figure 8: Comparison o Average Daily Energy Demand Due to Heat Flow Through an Extensive

Green versus Conventional Roo in Ottawa, Canada17

876

HeatFlow(kWh)543210 Kar

enLiu

Nov Dec Jan Feb Mar Apr May Jun Jul Aug SepThis chart shows the average daily energy demand due to observed heat ow through a green and conventional roo.

The period o evaluation was November 22, 2000, through September 30, 2001.

when the outdoor air temperature was

less than 55F (13C), the heat fux was

almost 50 percent less or the green

roo than or the conventional roo.21

Reduced Air Pollution and Greenhouse

Gas Emissions.As described in the

Trees and Vegetation chapter, vegetation

removes air pollutants and greenhouse

gas emissions through dry deposition and

carbon sequestration and storage. The

reduced energy demand rom green roos

also reduces air pollution and greenhouse

gas emissions associated with energy

production. Further, because ground-level

ozone orms more readily with the rise in

air temperatures, green roos help slow

the ormation o ground-level ozone by

lowering air temperatures. As with treesand vegetation, when selecting vegetation

or a green roo, building owners in areas

with poor air quality may want to consider

the volatile organic compound (VOC)

emissions rom certain plant species, as

VOCs are a ground-level ozone pre-cursor.

Plant suraces can remove certain pollut

ants rom the air through dry deposition.

A green roo can remove particulate mat

ter (PM) and gaseous pollutants, includ

ing nitrogen oxides (NOX), sulur dioxide

(SO2), carbon monoxide (CO), and ground-

level ozone (O3) rom the air. Many studies

have investigated the potential air pollutantremoval o green roos:

Researchers estimate that a 1,000-squareoot (93 m2) green roo can remove

about 40 pounds o PM rom the air in a

year, while also producing oxygen and

removing carbon dioxide (CO2) rom

the atmosphere.22 Forty pounds o PM

is roughly how much 15 passenger cars

will emit in a year o typical driving.23

A modeling study or Washington, D.C.,examined the potential air quality benets o installing green roos on 20

percent o total roo surace or buildings

with roos greater than 10,000 square eet

(930 m2). Under this scenario, green roos

would cover about 20 million square

eet (almost 2 million m2) and remove,

GREEN ROOFS DRAFT 7


11/29

annually, about 6.0 tons o O3 and almost

6 tons o PM o less than 10 microns

(PM10), or the equivalent o the pollutants

that could be absorbed by about 25,000

to 33,000 street trees.24

A similar study or the midtown areao Toronto modeled various greenroo scenarios and compared pollutant

reductions with existing baseline urban

tree and shrub benets. One scenario

involved green roos on fat roo surac

es, representing 20 percent o midtown

roos in total, such as commercial, high-

rise residential, and institutional build

ings. In that scenario, the green roos

removed about 10 to almost 20 percent

o the pollution that existing trees andshrubs remove, depending on the pol

lutant examined. I green roos were

added to all available suraces across

midtown Toronto, the model predicted

that green roos collective perormance

would increase to between roughly 25

and 45 percent o the reductions cur

rently obtained by existing vegetation.25

Vegetation and the growing medium on

green roos also can store carbon. Becausemany o the plants are small and the grow

ing medium layer is relatively thin, green

roos tend not to have as large a carbon

storage capacity as trees or urban orests.

Improved Human Health and Comort.

Green roos, by reducing heat transer

through the roo o a building, can improve

indoor comort and reduce heat stress as

sociated with heat waves. The use o cool

roos (see Cool Roo chapter) providessimilar indoor air temperature benets.

These improvements in building comort

can yield human health benets, particu

larly in non-air conditioned buildings.

Enhanced Stormwater Management

and Water Quality.Another key benet

o green roos is that they can reduce and

Figure 9: Green Roo on Seattle Public Library

Municipal buildings, such as this public library in

Seattle, have oten been used to demonstrate the

benets o green roos and the easibility o the

technology.

GreenRoofsforHealthyCities/w

ww.g

reenroofs

.org

slow stormwater runo in the urban en

vironment. The plants and growing me

dium o a green roo, in the same manner

as other natural suraces and vegetation,

absorb water that would otherwise become

runo. The amount o rainall retained by

a green roo will depend primarily on the

depth o the growing medium and may also

be aected by the roo slope. Studies have

shown that extensive roos will typically

capture between 50 and nearly 100 percento incoming rain, depending on the amount

o growing medium used, the density o

vegetation, the intensity o an individual

rainstorm, and the requency o local rain

events.26An intensive green roo, with thick

er layers o growing medium, will capture

more rainall under comparable conditions

than an extensive roo. Field study results

below help illustrate these ndings:

A North Carolina study o actual greenroo perormance ound that test green

roos reduced runo rom peak rainall

events by more than 75 percent and that

the roos temporarily stored and then

released, through evapotranspiration,

more than 60 percent o all rainall.27

A Canadian green roo demonstration



12/29

measured signicant reductions in

runo over a six-month period, with

steep reductions in ve o the six

months, and then lower reductions in

one month that had many large rain

events, which did not allow the grow

ing medium to dry out between events.

Overall, this project showed the green

roo reduced runo by more than 50

percent.28

A green roo demonstration projectin Portland, Oregon, examined runo

reductions over a 15-month period. In

that study, a green roo with about our

inches (10 cm) o growing medium

reduced runo by almost 70 percent.29

In addition, the authors noted that theretention rate appeared to increase over

time, which might be related to matur

ing vegetation. Because o the benets

in controlling stormwater, Portland has

approved green roos (or eco-roos)

as a technique to help meet stormwater

management requirements or new devel

opment and redevelopment projects. 30

Stormwater retention will vary with local

conditions, and communities generallyconsider this when projecting the poten

tial stormwater benets o green roos in

their area.

Even when a green roo does not retain all

the water rom a storm, it can detain runo

or later release and reduce the runo

rate. For example, the same Portland study

demonstrated that the green roo reduced

peak run-o rates by 95 percent during an

intense storm.31The North Carolina studyound that average peak runo rates rom

the green roos were roughly 75-85 percent

less than average peak rainall rates, so that

even when rain was alling on average at

Various research projects are un

derway to continue monitoring

pollutants in stormwater runoff

from green roofs, such as those at

Pennsylvania State Universitys Green

Roo Research Center, North Carolina

State Universitys Greenroo Research

program, the Green Roo Test Plots

research at the Chicago Center or

Green Technology, and Portland, Or

egons Eco-Roo program.

about 1.5 inches/hr (42 mm/hr), it ran o

the green roo at less than 0.25 inches/hr

(6 mm/hr).32 Reduced rates o runo can

help communities minimize fooding and

combined sewer overfow (CSO) events.*

The plants and growing medium o a

green roo not only retain and delay

the release o stormwater but also act

as a lter. Findings rom various studies

demonstrate the ability o green roos to

remove pollutants and highlight the need

to select growing media careully to avoid

elevated levels o certain pollutants, which

may initially leach rom organic materials.

A 2005 Canadian report synthesized past

studies on this issue.33 It noted that sev

eral studies rom Europe had ound that

green roos can bind and retain signicant

levels o pollutants, with one study stat

ing that green roos could remove up to

95 percent o the cadmium, copper, and

lead rom stormwater runo. The study

also summarized ndings rom a moni

toring program on a green roo in York,

Ontario, which ound decreased pollutant

concentrations compared to a control roo.

The reductions ranged rom 80 to almost

* Combined sewer systems are single-pipe systems that carr y sewage and stormwater runo together; when they overow during heavy rain, they dis

charge directly into surace waters.

GREEN ROOFS DRAFT 9


13/29

95 percent or several pollutants, such as

suspended solids, copper, and polycyclic

aromatic hydrocarbons. The same study,

however, ound increased concentrations

o nitrogen and phosphorous.

Recent research in Pennsylvania ound im

proved pH in green roo runo compared

to a conventional roo, as well as reduc

tions in total nitrate loadings based on the

reduced amount o stormwater rom the

green roo. The concentration o other pol

lutants in the green roo runo, in contrast,

was generally higher than concentrations

rom a conventional roo.34

As with the eld study in York, Ontario,

research in North Carolina ound increases

in total nitrogen and total phosphorous,

which the authors attributed to certain

compost materials in the roo substrate.35

Research in Portland and Toronto ound

that phosphorous levels appeared to de

crease over time as the green roo vegeta

tion matured and the phosphorous in the

initial substrate leached during rainall

events.36,37A German study also revealed

that a green roo retained more phosphate

as it matured, with retention percentage in

creasing rom about 26 percent in the rst

year to about 80 percent in the ourth.38

Enhanced Quality o Lie. Green roos

can provide many o the same quality o

lie benets as other urban greenery. People

in taller, neighboring buildings may enjoy

looking down at a rootop garden. Allowing

public access to rootop gardens provides

residents another green space to enjoy.

Finally, some researchers are evaluating the

potential or green roos to provide a sae

habitat or rare or endangered species, re

moving them rom ground-level predators.39

3.2 Costs

The costs o green roos vary depending

on the components, such as the growing

medium, type o roong membrane, drain

age system, use o encing or railings, and

type and quantity o plants. A 2001 reportestimated that initial costs start at $10

per square oot (0.09 m2) or the simpler,

extensive roo and $25 per square oot or

intensive roos.40 Other estimates assume

$15 to $20 per square oot. Costs in Germa

ny, where green roos are more prevalent,

range rom $8 to $15 per square oot.41

Prices in the United States may decline as

market demand and contractor experience

increase.

Initial green roo costs are more than those

o most conventional and cool roo tech

nologies (see Cool Roos chapter). Green

roos have a longer expected lie, though,

than most roong products, so the total

annualized costs o a green roo may be

closer to those o conventional and cool

roos. Los Angeles estimated that to retro

t a building with an extensive green roo

would cost rom $1.03-$1.66 per square

oot, on an annualized basis, while aconventional re-roong would range rom

$0.51-$1.74 per square oot.42

In addition to construction costs, a build

ing owner incurs maintenance costs to

care or the plants on a green roo. Al

though the level o care depends on plant

selection, most o the expenses arise in

the rst years ater installation, as the

plants establish themselves and mature.

For either an intensive or extensive roo,maintenance costs may range rom $0.75

to $1.50 per square oot. The costs o

maintaining an extensive roo decrease

ater the plants cover the entire roo,

whereas maintenance costs will remain

more constant or an intensive roo.43



14/29

3.3 Benet-Cost Considerations

Although a green roo might have higher

initial costs than most conventional or

cool roos, a ull lie-cycle analysis can

identiy how the roo benets the build

ing owner. In many cases, these benetsjustiy the cost o green roos in densely

populated areas. In addition, a building

owner can directly benet rom reduced

energy use, reduced stormwater manage

ment ees, and increased roo lie. Finally,

the widespread adoption o green roos

may provide signicant, indirect net ben

ets to the community.

Although ew detailed, ull lie-cycle analy

ses exist, researchers and communitiesare beginning to invest in these evalua

tions. A report on the use o green roos

in New York City outlined one ramework

or a cost-benet analysis o green roos.44

The ramework incorporates both private

and public benets and costs (see Table

1). Under most hypothetical scenarios,

a green roo project yields net benets

when assessed with public benets, such

as reduced temperature and stormwater.

Table 1: Benet-Cost Elements or Green Roos

Under a high-perormance scenario that

generally assumes reduced costs rom

widespread adoption o green roo tech

nology and a mature market, an owner

would achieve net benets based on pri

vate benets alone.

A University o Michigan study compared

the expected costs o conventional roos

with the cost o a 21,000-square-oot (1,950

m2) green roo and all its benets, such

as stormwater management and improved

public health rom the NOXabsorption. The

green roo would cost $464,000 to install

versus $335,000 or a conventional roo

in 2006 dollars. However, over its lietime,

the green roo would save about $200,000.

Nearly two-thirds o these savings would

come rom reduced energy needs or the

building with the green roo.45

Portland, Oregon, meanwhile, has begun a

comprehensive cost-benet analysis o its

current eco-roo program, as the city plans

to expand green roo coverage rom 6

acres (24,300 m2) in 2007 to over 40 acres

(162,000 m2) in 2012.46

Benets/Costs Energy, Hydrology, and UHI Benets Other Benets

Private Benets

Reduced energy use

Extended service lie

Noise reductionAesthetic valueFood production

Public Benets

Reduced temperature

Reduced stormwater

Reduced installation costs (rom widespread

technology use)

Reduced air pollutantsReduced greenhouse gasesHuman health benets

Private Costs

Installation

Architecture/Engineering

Maintenance

N/A

Public Costs Program administration N/A

GREEN ROOFS DRAFT 11


15/29

Cool and Green Roos: Diferent Options or

Diferent Motivations

Cool and green roos both help to mitigate urban heat islands. The two technologieshave dierent cost and perormance implications, though, and the motivations or

selecting one or the other are typically dierent.

Cool roos generally have a minimal incremental

cost compared to their conventional equivalent.

Depending on the type o product (e.g., asphalt

shingle, concrete tile), costs can range roughly

between $0.50 to $6.00 per square oot. Costs can

vary greatly, though, depending on the size o the

job, ease o access to the roo, and local market

actors. The initial cost o a green roo, on the otherhand, is much higher, starting rom $10 per square

oot or the basic, extensive green roo.

Both cool and green roos lower surace and air temperatures and reduce summer

time peak and overall energy demand. The extent o the energy savings varies de

pending on actors including the local climate, attic ventilation and insulation levels,

andparticularly or green roosthe design and

maintenance o the roo.

Green roos provide additional benets, including

reducing and ltering stormwater runo, absorbingpollutants and CO2, providing natural habitat and

a sound barrier, and potentially serving as a recre

ational green space and having aesthetic value.

Communities or building owners with limited

budgets, who are primarily interested in energy savings or reducing peak energy

demand, generally ocus on cool roos. Whereas others, who can consider lie-cycle

costs and public benets, and who are interested in broader environmental impacts,

particularly improving stormwater management, may choose to install green roos.

Sustainability leaders, such as Chicago, recognize the value and opportunity or

both cool and green roo technologies and are supporting eorts to encourage

both options.



16/29

Energy Savings and

Green Roos

For building owners and communities primarily interested in saving

energy, cool roos and other energy

eciency measures are generally

more cost-eective than green roos.

(See the Cool Roo chapter and the

ENERGY STAR website

or inorma

tion about a wide array o cost-e

ective energy ecient products and

practices.) Green roos provide ben

ets beyond energy savings, though,which is why they are attractive to

diverse interest groups and sustain-

ability advocates.

Figure 10: Green, or Eco, Roo in

Portland, Oregon

This apartment building in Portland, Oregon, is

among the 6 acres (24,300 m2) o green roos in the

city, as o 2007. Many roos remain candidates to

become green roos.

PortlandBureauofEnvironmentalServices

4 Other Factors to Consider

4.1 Site Characteristics

Recommendations or ideal site charac

teristics vary and oten depend on project

or program objectives. For example, Chi

cago and New York City are ocusing on

hot spot areas, which are oten ound in

dense, built up urban cores. Green roos

may be the only option to provide an e

ective amount o vegetation in these older

city centers that have vast amounts o

impervious cover and ew opportunities to

retroactively plant shade vegetation. Fur

ther, entities interested in providing recre

ational space or improving aesthetics may

also ocus on high density areas that arevisible rom adjoining or near by buildings.

On the other hand, stakeholders ocused

on saving energy and managing stormwater

oten target low-to-medium rise buildings

that have a large roo area. These sites, such

as the Fords Dearborn Truck Plant in Michi

gan, may be ound in less developed areas.

From a structural standpoint, existing roos

with concrete structural systems likely will

require the least amount o intervention;

roos with steel deck can require the most.

Installing a green roo on a fat or low-

sloped roo generally will be easier than

installing one on a steep-sloped roo. Also,

green roos tend to be easier to design into

new rather than existing buildings, given

that loads and other requirements can be

included in the design process. However,

retrot installations are becoming increas

ingly common in the expanding greenroo market. Many existing buildings, such

as low-sloped residential and commercial

buildings with large roo areas, can be

modied without signicant disruption

when replacing an old roo. For example,

projects at Carnegie Mellon University,

Tobyhanna Army Depot, and the Albemarle

http://www.energystar.gov/http://www.energystar.gov/


17/29

The Green Roo Continuum

The decision to install an extensive

or intensive green roo depends onavailable resources and the building

owners goals or how the roo will

be used. For example, someone with

a limited budget who desires minimal

maintenance and is mainly interested

in the energy and environmental ben

ets o a green roo, would most like

ly install an extensive green roo. On

the other hand, someone who wants

to create an accessible garden and

is able to maintain the green space,will probably install a more intensive

green roo. Many green roos incorpo

rate a combination o extensive and

intensive green roo eatures. These

semi-extensive or semi-intensive

green roos lie within the continuum

o green roo types, with extensive

and intensive at each end o the

spectrum.

County, Virginia, oce building have high

lighted the ease o replacing stone-ballast

on existing roos with vegetative layers.47

4.2 Installation and Maintenance

Whether extensive, intensive, or somewhere in between, green roos generally

consist o the same basic components.48

From the top layer down (see Figure 12),

these include:

Vegetation.The choice o vegetationdepends on the type o roo (extensive

or intensive), building design, local

climate, available sunlight, irrigation

requirements, anticipated roo use, and

similar actors: Extensive green roo plants are

typically hardy perennials. They are

preerably shallow-rooting, sel-

generating plants that spread rap

idly and require minimal nutrients.

They should tolerate sun, wind,

and extreme temperature fuctua

tions. Succulents, such as sedums,

are well adapted or green roos

because they are drought-resistant

and their high water content makesthem re resistant. Sedums come

Figure 11: A Green Roo Replaces a Stone Ballast Roo

Albemarle County, Virginia, replaced the stone ballast

roo on its county ofce building with a green roo in

2005. The project received money rom the Chesapeake

Bay Program through the Virginia Department o

Conservation and Recreation.

GregorPatsch/AlbemarleCounty



18/29

Figure 12: Typical Layers o a Green Roo

KarenLiu/NationalResearchCouncilofCanada

in a wide variety o sizes, textures,

and colors. Building owners also

can ensure that the selected plants

suit USDA plant hardiness zones or

their area.49

Intensive green roos have deeper

growing media, which allows them

to incorporate larger plants, in

cluding shrubs, bushes, and trees,

in their design. Most intensivegreen roos also have irrigation

systems that can support a wide

variety o plants.

A lightweight, engineeredgrowing me-dium may or may not include soil as

the primary organic matter. The plant

ing media used in green roo systems

are usually engineered to provide the

best support or plants with the lightest

weight and can be tailored to maximize

water retention without water-logging

the plants. A growing medium should

ideally last as long as the roo it will

cover. Typically, the growing medium

will consist primarily o lightweight

inorganic mineral materials (at least 80

percent) and up to 20 percent organic

materials like topsoil.50 Extensive green

roos use up to roughly 6 inches (15

cm) o growing medium51 while inten

sive green roos use 8 inches (20 cm)

or more.52

Aflter membrane is usually a geotextile that allows excess water rom

the growing medium to fow out, while

preventing the ne particles rom wash

ing away and clogging the roo drain.

Adrainage layerhelps the excess water rom the growing medium to fow tothe roo drain, which prevents over

loading the roo and provides a good

air-moisture balance in the growing

medium. Some drainage layers take the

orm o egg crates to allow or some

water storage.

Aroot barriercan protect the roomembrane rom aggressive plant roots,

which may penetrate the waterproong

layer and cause leaks.

Awaterproofng/roofng membraneprotects the building rom water pen

etration. Any roong membrane can be

used in green roos, although single-ply

waterproong membranes are gener

ally thicker and more durable on green

roos than on conventional ones. Some



19/29

membranes are naturally protected

rom root penetration, while others

require a root barrier.

Acover boardis a thin, semi-rigid boardthat provides protection, separation, and

support or a roong membrane. Thermal insulation can be installed

either above or below the membrane

o a green roo. The insulation value o

the growing medium in a green roo

increases as its moisture content de

creases. However, green roos are not

a substitute or conventional insula

tion; using the recommended insula

tion levels or ones local climate helps

conserve energy.

Avapor barrieris typically a plastic oroil sheet that resists passage o mois

ture through the ceiling.

Building and roo structural sup-port.The components o a green roo

weigh more than conventional roong

materials, and thus the roo requires

support panels. Not only are the roo

ing membranes and other materials

heavier on a green roo, but the roo

design also must account or the weighto water-saturated plants and growing

medium. An extensive roo typically

weighs rom 15-30 pounds per square

oot, although the range will depend on

the depth o the growing medium and

other site-specic actors.53An inten

sive roo can weigh much more, with

signicantly greater depth o growing

medium, more extensive vegetation, and

people using the space. Building own

ers must ensure that the structure cansupport the green roo even when ully

saturated, in addition to meeting build

ing code requirements or snow and

wind loads. Reinorcing roo supports

on existing buildings adds to the proj

ect cost but can usually be worked into

building retrot or renovation plans. It

is oten easier to put green roos on new

Inverted Roo Membrane

Assemblies (IRMAs) and

Green Roos

Inverted roo membrane assemblies

(IRMAs) have insulation above the

waterproong membrane, as opposed

to conventional roos, which have

insulation below the membrane. This

design protects the membrane and

prolongs the lie o the roo. A green

roo that has insulation between its

vegetative layer and the waterproong

membrane is an IRMA, with the veg

etation protecting the membrane andweighing down the insulation. More

conventional IRMAs use concrete pav

ers or stones or ballast. These IRMAs

oten make good candidates or green

roo retrots, as the conventional bal

last can be replaced with the neces

sary green roo layers.

buildings, as the requirements or the

added roo load can be included as parto the initial design parameters, and the

cost or the upgrade is usually minimal.

Although both extensive and intensive

green roos share these basic components,

their characteristics vary (see Figure 13).

Most important, the intensive green roos

are likely to require more structural sup

port and enhanced irrigation systems to

support the wider variety o plants, in

creased weight loads, and desired publicaccess. However, intensive roos will prob

ably also retain more rainall and support

more species.54

In addition, any green roo generally will

require some ongoing maintenance. Ex

tensive green roos not designed or pub

lic access have ewer obligations. For an



20/29

intensive roo, maintenance can be continu

ous, similar to a traditional garden, because

aesthetics will be more important.

For either roo, early weed control is im

portant to ensure that the installed plants

have a chance to spread and to minimize

the opportunity or invasive weeds to

take root. According to a ederal guide on

green roos,55 weeding might be necessary

monthly or quarterly or the rst two years

and might be reduced to only once a year

in many cases ater the plants have ully

covered the roo. The guide also lists other

important maintenance activities including:

Fertilize. Given the thin layer ogrowing medium, building owners or

managers might need to apply a slow

release ertilizer once a year to avoid

soil acidity, especially when the plants

are rst establishing themselves.

Irrigate.An ideal green roo couldrely on natural irrigation, especially or

extensive roos. However, some green

roos might require irrigation based

on local climate and the stage o plant

growth or a particular project. Irriga

tion might also be needed to reduce

re risks or to increase evaporative

cooling. Almost all intensive green

roos need irrigation systems. Extensive

green roos, however, may only need

them during plant establishment. For

large, extensive green roos, building

owners oten install a drip irrigation

system, which is generally inexpensive

and saves the time and eort o having

someone manually water the roo. Replant. Over time, some level o

replanting or addition to the growing

medium might be necessary.

Figure 13: Comparison o Common Features o Extensive and Intensive Roos

Extensive vs. Intensive

(Low-Prole/Ecoroos) (High-Prole/Roo Gardens)

Low growth media: 2 6" > 6"-15" and deeper Lightweight: 13 50 lbs/s Heavier weights: 50+ lbs/s Low growing plants: 1" 24" H Trees, shrubs and more Less variety o plants: Alpine types, Huge variety o plant selection/

succulents, herbs, some grasses architectural eatures depending

and mosses on loads, design & budget

Usually non-accessible and Designed or human recreation:non-recreational gardening, socializing, etc.

Slopes up to 30 & higher Relatively fat Less expensive: $5-$25/s More expensive: $25-$40+/s Low water requirements Irrigation usually necessary Low maintenance Higher maintenance

LindaVelasquez

There is no ofcial denition o an extensive or intensive green roo. This chart is not meant to strictly dene these

green roo types and instead aims to describe the general characteristics o roos at each end o the continuum.



21/29

Figure 14: A Modular Green Roo on a

Sloped Residential Roo

This home in Arizona shows a modular green roo

on a steep-sloped roo.

Greenroofs

.comandKellyLuckett/Gr

eenRoofBlocks

Modular Green Roo Systems

Some green roo systems use modular

components. These components are

generally plastic trays a ew eet long

(~0.5-1 m) on each side and several

inches (~10-20 cm) deep. They are

lled like fowerpots with growing

media and the desired plants and

placed directly on top o the existing roo. The grid o trays covers the

roos surace to provide benets simi

lar to built-in green roos. Moving or

replacing individual modules is poten

tially easier than changing or repair

ing parts o a non-modular green roo.

Modular roos, however, are relatively

new, and have not been as widely

studied as non-modular roos.

Clean Gutters. Similar to conventionalroos, clean gutters decrease the risk o

standing water and leaks. It is also nec

essary to keep drains and gutters clear

o plant growth to prevent blockage.

In addition to routine maintenance, green

roos may require repairs over time, al

though the expected lie o a green roo is

about twice that o a conventional roo.56

I correctly installed, the membrane under

the vegetation o a green roo is expected

to last 30 to 50 years.

4.3 Fire Saety

Green roos, when saturated with water,

can retard the spread o re,57 but dry

plants on a green roo can be a re hazard.

The most common ways to increase re

saety are to:

Avoid grasses and plants that coulddry up in summer and instead use reresistant plants, like sedums, and a

growing medium that is low in organic

material content.

Construct re breaks on the roo2-oot (0.6 m) widths o concrete or

gravel at 130-oot (40 m) intervals.

Another precaution that some practitioners

recommend is to install sprinkler irrigation

systems and connect them to a re alarm.

5 Green Roo Initiatives

Green roo research eorts are growing with

an increasing number o universities oer

ing courses or developing centers ocused

on improving our understanding o green

roo technology. Many communities are also

taking action by encouraging or sponsor

ing green roo projects. These initiatives are

typically motivated by various environmental

concerns, mainly stormwater management,but also the desire to reduce urban heat

islands and enhance the urban ecosystem.

Many o these eorts involve a single dem

onstration or showcase project as a highly

visible means to promote green roo technol

ogy, such as the green roo on Atlantas City



22/29

Many green roo projects are motivated

not by government policies but by a de

sire to show a commitment to sustain-

able design and the environment.

Hall patio. Some cities such as Chicago, Port

land, Seattle, and Toronto have been develop

ing more coordinated programs and policies

to promote green roos. The Heat Island

Reduction Activities chapter provides many

examples demonstrating the wide range o

green roo eorts. Table 2 identies some o

the research activity and options available or

taking action to advance green roos.

Green building programs in many com

munities provide another opportunity to

encourage green roo installation. The U.S.

Green Building Council (USGBC) Leader

ship in Energy and Environmental Design

(LEED) Rating System (see ) and Green Globes operated by the

Green Building Initiative (GBI) in the

United States (see ), are

two rating systems that communities are

using. These and other systems give credit

or a broad range o building and devel

opment techniques that save energy and

protect the environment. Green roos can

achieve credit under multiple categories

such as stormwater management, heat

island mitigation, water eciency, energy

and atmosphere, materials and resources,

and innovation and designdepending onhow they are constructed.

Figure 15: A Newly Installed Green Roo in New York City

2008J.BurlingChaseforSustainableSouthBronx

Initiatives to install green roos in urban areas reduce urban heat islands and can help to create jobs in the local

economy, such as this roo installed by graduates o Sustainable South Bronxs Bronx Environmental Stewardship

Training (BEST) program.

http://www.usgbc.org/http://www.usgbc.org/http://www.thegbi.org/http://www.thegbi.org/http://www.usgbc.org/http://www.usgbc.org/


23/29

Table 2: Examples o Green Roo Initiatives

Type o Initiative Description Links to Examples

Research University

programs

- Michigan State University began its

Green Roo Research Program in 2000 to assist with the design and study o

Fords Rouge Plant. The program has since expanded and now investigates

green roo plant selection among other topics.

- Penn State

Universitys Center or Green Roo Research studies the energy savings,

stormwater retention and ltration, and other benets o green roos.

- North Carolina State University has

extensive green roo test sites in Goldsboro and Kinston, North Carolina, as

part o the Biological and Agricultural Engineering Program.

- The University o Central Florida ocuses

primarily on stormwater management, which has led to its investigations o

green roos.

- The British Columbia Institute o

Technologys Centre or the Advancement o Green Roo Technology collabo

rates with industry to support and improve the deployment o green roos.

Voluntary eorts Demonstration

projects

- For background inormation on Chicago

City Halls green roo, see the Links section o this site.

This site provides an overview o the Atlanta City Hall green roo demon

stration project.

Incentives - Chicago has sponsored a green roo

grant program or several years. Grants o up to $5,000 each were available

in the application cycle that ended in January 2008. See the Department oEnvironment page and browse under Initiatives and Programs.

- Torontos green

roo incentive program oers grants o Canadian $50 per square meter

or eligible projects, up to a total o $10,000 or single-amily homes and

$100,000 or all other buildings.

http://www.hrt.msu.edu/greenroofhttp://hortweb.cas.psu.edu/research/greenroofcenterhttp://www.bae.ncsu.edu/greenroofshttp://www.stormwater.ucf.edu/http://commons.bcit.ca/greenroof/http://www.chicagogreenroofs.org/http://www.atlantaga.gov/mayor/energyconservationgreenroof.aspxhttp://egov.cityofchicago.org/http://www.toronto.ca/greenroofs/incentiveprogram.htmhttp://www.toronto.ca/greenroofs/incentiveprogram.htmhttp://egov.cityofchicago.org/http://www.atlantaga.gov/mayor/energyconservationgreenroof.aspxhttp://www.chicagogreenroofs.org/http://commons.bcit.ca/greenroof/http://www.stormwater.ucf.edu/http://www.bae.ncsu.edu/greenroofshttp://hortweb.cas.psu.edu/research/greenroofcenterhttp://www.hrt.msu.edu/greenroof


24/29

Table 2: Examples o Green Roo Initiatives (Continued)

Type o Initiative Description Links to Examples

Voluntary eorts Incentives - Portland, Or

egon, oers grants, workshops, and other technical assistance to support

green roos.

- The Hous

ton Downtown Management District (HDMD) Vertical Gardens Matching

Grant initiative is intended to assist in the acilitation o wall cover plantings

and exceptional landscaping on blank walls, parking garages, and side

walks; improving overall aesthetics, pedestrian comort, and air quality; and

reducing the heat island eect.

Outreach &

education

- EPAs Heat Island Reduction Initiative pro

vides inormation on the temperature, energy, and air quality impacts rom

green roos and other heat island mitigation strategies.

- EPAs Oce o Water highlights design options, including green roos, that reduce

stormwater runo and water pollution.

- Green Roos or Healthy Cities hosts a series o

green roo design and implementation workshops throughout North America.

Policy eorts Density bonus

provisions in

zoning codes

- Document that highlights eorts o Chicago; Seattle; Portland, Or

egon; Toronto; and Waterloo, Ontario, to encourage green roo installations

by oering density bonus incentives in their zoning codes.

http://www.portlandonline.com/bes/index.cfm?c=43077http://www.houstondowntown.com/Home/Business/DoingBusiness/DevelopmentAssistance/Development%20Assistance.PDFhttp://www.houstondowntown.com/Home/Business/DoingBusiness/DevelopmentAssistance/Development%20Assistance.PDFhttp://www.houstondowntown.com/Home/Business/DoingBusiness/DevelopmentAssistance/Development%20Assistance.PDFhttp://www.epa.gov/heatisland/http://cfpub.epa.gov/npdes/home.cfm?program_id=298http://www.greenroofs.org/http://commons.bcit.ca/greenroof/publications/2006_regulations.pdfhttp://commons.bcit.ca/greenroof/publications/2006_regulations.pdfhttp://commons.bcit.ca/greenroof/publications/2006_regulations.pdfhttp://commons.bcit.ca/greenroof/publications/2006_regulations.pdfhttp://commons.bcit.ca/greenroof/publications/2006_regulations.pdfhttp://www.greenroofs.org/http://cfpub.epa.gov/npdes/home.cfm?program_id=298http://www.epa.gov/heatisland/http://www.houstondowntown.com/Home/Business/DoingBusiness/DevelopmentAssistance/Development%20Assistance.PDFhttp://www.portlandonline.com/bes/index.cfm?c=43077


25/29

6 Resources

Table 3 lists some guidance documents and

organizations that promote green roos.

Table 3: Green Roo Resources

Name Description Web Link

Guidance Documents

U.S. Department o Energy

Federal Technology Alert:

Green Roos

DOEs Energy Efciency and Renewable Energy pro

gram publishes technology alerts and developed

this primer on green roo technology.

Green Roos as Urban

Ecosystems: Ecological

Structures, Functions, and

Services

The journal Bioscience November 2007 issue contains

this comprehensive article summarizing the research

on green roos and their costs and benets.

National Roong Contrac

tors Association Green

Roo Systems Manual

The NRCA has recently released a guidebook or

sale that ocuses on the waterproong needs o

green roos.

Los Angeles Green Roo

Resources Guide

The City o Los Angeles developed this guide as a

resource or individuals and groups interested in

developing green roos in Los Angeles. This guide

includes inormation on how to plan, design, and

maintain a green roo.

Other Resources

Green Roos or Healthy

Cities

Green Roos or Healthy Cities oers resources

on green roo installation, benets, projects, andtraining. This group also publishes the Green Roof

Infrastructure Monitor.

Greenroos.com Greenroos.com provides green roo industry

resources, including how-tos, plant lists, reerences,

and an international database o green roo projects.

Chicago Green Roo

Program

Chicagos Green Roo Program has online inorma

tion on building green roos in Chicago, including

an aerial map o completed and planned projects,

requently asked questions, eatured projects, and

links to other resources.

http://www.nrel.gov/docs/fy04osti/http://www.aibs.org/bioscience-press-releases/resources/11-07.pdfhttp://www.aibs.org/bioscience-press-releases/resources/11-07.pdfhttp://www.aibs.org/bioscience-press-releases/resources/11-07.pdfhttp://www.nrca.net/rp/pubstore/http://www.fypower.org/pdf/LA_GreenRoofsResourceGuide.pdfhttp://www.fypower.org/pdf/LA_GreenRoofsResourceGuide.pdfhttp://www.fypower.org/pdf/LA_GreenRoofsResourceGuide.pdfhttp://www.greenroofs.org/http://www.greenroofs.com/http://www.chicagogreenroofs.org/http://www.chicagogreenroofs.org/http://www.greenroofs.com/http://www.greenroofs.org/http://www.fypower.org/pdf/LA_GreenRoofsResourceGuide.pdfhttp://www.fypower.org/pdf/LA_GreenRoofsResourceGuide.pdfhttp://www.fypower.org/pdf/LA_GreenRoofsResourceGuide.pdfhttp://www.nrca.net/rp/pubstore/http://www.aibs.org/bioscience-press-releases/resources/11-07.pdfhttp://www.aibs.org/bioscience-press-releases/resources/11-07.pdfhttp://www.aibs.org/bioscience-press-releases/resources/11-07.pdfhttp://www.nrel.gov/docs/fy04osti/


26/29

Endnotes

1 Rose, L.S., H. Akbari, and H. Taha. 2003. Characterizing the Fabric o the Urban Environment: A

Case Study o Greater Houston, Texas. LBNL-51448, January 2003.

2 Liu, K. and B. Baskaran. 2003. Thermal perormance o green roos through eld evaluation.

National Research Council o Canada-46412.

3 Final Report, Green Roo Industry Survey, 2006, prepared by Green Roos or

Healthy Cities, April 2007. Retrieved 7 Dec. 2007 rom .

4 The Green Roo Research Program at MSU. Retrieved 7 August 2007 rom .

5 Peck, S.W., C. Callaghan et al. 1999. Greenback, rom Green Roos: Forging a New Industry in

Canada. Prepared or Canada Mortgage and Housing Corporation.

6 Huang, J., H. Akbari, and H. Taha. 1990. The Wind-Shielding and Shading Eects o Trees on

Residential Heating and Cooling Requirements. ASHRAE Winter Meeting, American Society o

Heating, Rerigerating and Air-Conditioning Engineers. Atlanta, Georgia.

7 Akbari, H., D. Kurn, S. Bretz, and J. Hanord. 1997. Peak power and cooling energy savings oshade trees. Energy and Buildings. 25:139-148.

8 Sandier, S. and B. Givoni. 2002. Thermal Eects o Vines on Wall TemperaturesComparing

Laboratory and Field Collected Data. SOLAR 2002, Proceedings o the Annual Conerence o the

American Solar Energy Society. Reno, NV.

9 Department o Energy 2004. Federal Technology Alert: Green Roos. DOE/EE-0298, Washington, D.C.

10 Cummings, J., C. Withers, J. Sonne, D. Parker, and R. Vieira. 2007. UCF Recommissioning, Green

Roong Technology, and Building Science Training; Final Report. FSEC-CR-1718-07. Retrieved

18 December 2007 rom .

11 Liu, K. and B. Bass. 2005. Perormance o Green Roo Systems. National Research Council

Canada, Report No. NRCC-47705, Toronto, Canada.12 Rosenzweig, C., W. Solecki et al. 2006. Mitigating New York Citys Heat Island with Urban For

estry, Living Roos, and Light Suraces. Sixth Symposium on the Urban Environment and Forum

on Managing our Physical and Natural Resources, American Meteorological Society, January 31,

2006, Atlanta, GA.

13 Bass, B. and B. Baskaran. 2003. Evaluating Rootop and Vertical Gardens as an Adaptation Strate

gy or Urban Areas. National Research Council Canada, Report No. NRCC-46737, Toronto, Canada.

14 McKeough, T. Room to Improve. New York Times. 21 Feb. 2008. Retrieved 10 Mar. 2008 rom

.

15 Ford Motor Company. Ford Installs Worlds Largest Living Roo on New Truck Plant. Retrieved 2

August 2007 rom .

16 Liu, K. 2002. A National Research Council Canada Study Evaluates Green Roo Systems Thermal

Perormances. Proessional Roong.



18 Department o Environment. Chicago City Hall green roo project. Retrieved 18 October 2007

rom .

http://www.greenroofs.org/storage/2006grhcsurveyresults.pdfhttp://www.greenroofs.org/storage/2006grhcsurveyresults.pdfhttp://www.hrt.msu.edu/faculty/Rowe/Green_roof.htmhttp://www.hrt.msu.edu/faculty/Rowe/Green_roof.htmhttp://www.fsec.ucf.edu/en/publications/pdf/FSEC-CR-1718-07.pdfhttp://www.nytimes.com/2008/02/21/garden/21room.htmlhttp://media.ford.com/newsroom/release_display.cfm?release=15555http://egov.cityofchicago.org/http://egov.cityofchicago.org/http://media.ford.com/newsroom/release_display.cfm?release=15555http://www.nytimes.com/2008/02/21/garden/21room.htmlhttp://www.fsec.ucf.edu/en/publications/pdf/FSEC-CR-1718-07.pdfhttp://www.hrt.msu.edu/faculty/Rowe/Green_roof.htmhttp://www.hrt.msu.edu/faculty/Rowe/Green_roof.htmhttp://www.greenroofs.org/storage/2006grhcsurveyresults.pdfhttp://www.greenroofs.org/storage/2006grhcsurveyresults.pdf


27/29



20 Cummings, J., C. Withers, J. Sonne, D. Parker, and R. Vieira. 2007. UCF Recommissioning, Green

Roong Technology, and Building Science Training; Final Report. FSEC-CR-1718-07. Retrieved

18 Dec. 2007 rom .

21

Sonne, J. Energy Perormance Aspects o a Florida Green Roo, Fiteenth Symposium on Improving Building Systems in Hot and Humid Climates, July 24-26, 2006 Orlando, FL.

22 Peck, S. and M. Kuhn. 2003. Design Guidelines or Green Roos. Canada Mortgage and Housing

Corporation, Ottawa, and the Ontario Association o Architects, Toronto.

23 This comparison assumes each car will produce 0.1g o PM per mile (based on new ederal

standards that would limit PM emissions to this level or lower in passenger vehicles), and that

each car is driven 12,500 miles (20,000 km) in a year, which was the average mileage or a car

in America in 2004. See U.S. Department o Transportation Federal Highway Administration.

Annual Vehicle Distance Traveled in Miles and Related Data-2004. Highway Statistics 2004.

October 2005. Retrieved October 19, 2007 rom .

24 Casey Trees Endowment Fund and Limno-Tech, Inc. 2005. Re-Greening Washington, D.C.: A GreenRoo Vision Based on Quantiying Storm Water and Air Quality Benets. Washington, D.C.

25 Currie, B.A. and B. Bass. 2005. Estimates o Air Pollution Mitigation with Green Plants and

Green Roos Using the UFORE Model. Sixth Biennial Canadian Society or Ecological Economics

(CANSEE) Conerence, October 27-29, 2005, Toronto, Canada.

26 VanWoert, N.D., D.B. Rowe, J.A. Andresen, C.L. Rugh, R.T. Fernandez, and L. Xiao. 2005. Green

Roo Stormwater Retention: Eects o Roo Surace, Slope, and Media Depth. Journal o Envi

ronmental Quality 34:1036-1044.

27 Moran, A., B. Hunt et al. 2004. A North Carolina Field Study to Evaluate Greenroo Runo

Quantity, Runo Quality, and Plant Growth. Paper Presented at Green Roos or Healthy Cities

Conerence, Portland, OR, June 2004.

28 Liu, K. 2003. Engineering perormance o rootop gardens though eld evaluation. National

Research Council Canada, Report No. NRCC-46294, Ontario, Canada.

29 Hutchinson, D., P. Abrams et al. 2003. Stormwater Monitoring Two Ecoroos in Portland, Or

egon, USA. Proceedings o Greening Rootops or Sustainable Communities, 2003, Chicago, IL.

30 Portland. 2002. City o Portland EcoRoo Program Questions and Answers. Bureau o Environ

mental Services, Oce o Sustainable Development, City o Portland, Oregon, PL 0203, Port

land, OR.






33 Banting, D., H. Doshi, J. Li, and P. Missios. 2005. Report on the Environmental Benets and

Costs o Green Roo Technology or the City o Toronto. Department o Architectural Science,

Ryerson University.

34 Berghage, R., D. Beattie, A. Jarrett, and T. OConner. 2007. Greenroo Runo Water Quality. Fith

Annual Greening Rootops or Sustainable Communities Conerence, April 29-May 1, 2007.

http://www.fsec.ucf.edu/en/publications/pdf/FSEC-CR-1718-07.pdfhttp://www.fhwa.dot.gov/policy/ohim/hs04/htm/vm1.htmhttp://www.fhwa.dot.gov/policy/ohim/hs04/htm/vm1.htmhttp://www.fhwa.dot.gov/policy/ohim/hs04/htm/vm1.htmhttp://www.fhwa.dot.gov/policy/ohim/hs04/htm/vm1.htmhttp://www.fsec.ucf.edu/en/publications/pdf/FSEC-CR-1718-07.pdf


28/29






37

Van Seters, T., L. Rocha, and G. MacMillan. 2007. Evaluation o the Runo Quantity and QualityPerormance o an Existing Green Roo in Toronto, Ontario. Fith Annual Greening Rootops or

Sustainable Communities Conerence, April 29-May 1, 2007.

38 Kohler, M. and M. Schmidt. 2003. Study on Extensive Green Roos in Berlin. Translated by S.

Cacanindin. Retrieved 27 April 2006 rom .

39 Banting, D., H. Doshi, J. Li, and P. Missios. 2005. Report on the Environmental Benets and

Costs o Green Roo Technology or the City o Toronto. Department o Architectural Science,

Ryerson University.

40 Peck, S. and M. Kuhn. 2001. Design Guidelines or Green Roos. National Research Council

Canada, Toronto, Canada.

41 Scholz-Barth, K. 2001. Green Roos: Stormwater Management rom the Top Down. Environmen

tal Design & Construction.

42 City o Los Angeles, Environmental Aairs Department. 2006. Green RoosCooling Los Ange

les (A Resource Guide). Los Angeles, CA.



44 Rosenzweig, C., S. Gan, and L. Parshall (Eds.). 2006. Green Roos in the New York Metropoli

tan Region: Research Report. Columbia University Center or Climate Systems Research and

NASA Goddard Institute or Space Studies. New York. 59 pages.

45 Clark, C., P. Adriaens, and F.B. Talbot. 2007. Green Roo Valuation: A Probabilistic Analysis o

Environmental Benets.

46 Personal correspondence with Tom Liptan, Portland Bureau o Environmental Services, 18 De

cember 2007.

47 In 2005, Carnegie Mellon University replaced a stone ballast roo on Hamerschlag Hall with a

green roo . When Albemarle County

replaced a stone ballast roo with a green roo in 2007, it did not have to modiy the roo be

cause the saturated vegetative layer weighed about the same as the stones and the underlying

membrane and insulation remained the same . In 2006, Tobyhanna Army Depot replaced a stone ballast roo

with a modular green roo. The roo had already been designed with capacity to support an

extra foor, so no modication was required to install a green roo .



49 Department o Energy. 2004. Federal Technology Alert: Green Roos. DOE/EE-0298, Washington, D.C.

50 Beattie, D., and R. Bergharge. 2004. Green Roo Media Characteristics: The Basics. In Greening

Rootops or Sustainable Communities, Portland, Oregon, June 2004.

51 Scholz-Barth, K. 2001. Green Roos: Stormwater Management rom the Top Down. Environmen

tal Design & Construction.

http://www.roofmeadow.com/http://www.greenroofs.com/projects/pview.php?id=292http://www.albemarle.org/department.asp?department=planning&relpage=8660http://www.albemarle.org/department.asp?department=planning&relpage=8660http://aec.army.mil/usaec/publicaffairs/update/win07/win0709.htmlhttp://aec.army.mil/usaec/publicaffairs/update/win07/win0709.htmlhttp://aec.army.mil/usaec/publicaffairs/update/win07/win0709.htmlhttp://aec.army.mil/usaec/publicaffairs/update/win07/win0709.htmlhttp://www.albemarle.org/department.asp?department=planning&relpage=8660http://www.albemarle.org/department.asp?department=planning&relpage=8660http://www.albemarle.org/department.asp?department=planning&relpage=8660http://www.greenroofs.com/projects/pview.php?id=292http://www.roofmeadow.com/


29/29



54 Coman, R. Vegetated Roof Systems: Design, Productivity, Retention, Habitat, and Sustainability

in Green Roof and Ecoroof Technology. (Doctoral Dissertation, The Ohio State University, 2007.)


56 See, e.g., Department o Energy (2004). Federal Technology Alert: Green Roos. DOE/EE-0298,

Washington, D.C.; and City o Los Angeles, Environmental Aairs Department (2006). Green

RoosCooling Los Angeles (A Resource Guide). Los Angeles, CA.

