Energy Efficiency (2010) 3:53–109
DOI 10.1007/s12053-008-9038-2
Potential benefits of cool roofs on commercial buildings:
conserving energy, saving money, and reducing emission
of greenhouse gases and air pollutants
Ronnen Levinson & Hashem Akbari
Received: 5 December 2007 / Accepted: 17 December 2008 / Published online: 14 March 2009
# Springer Science + Business Media B.V. 2009
Abstract Cool roofs—roofs that stay cool in the sun
by minimizing solar absorption and maximizing
thermal emission—lessen the flow of heat from the
roof into the building, reducing the need for space
cooling energy in conditioned buildings. Cool roofs
may also increase the need for heating energy in cold
climates. For a commercial building, the decrease in
annual cooling load is typically much greater than the
increase in annual heating load. This study combines
building energy simulations, local energy prices, local
electricity emission factors, and local estimates of
building density to characterize local, state average,
and national average cooling energy savings, heating
energy penalties, energy cost savings, and emission
reductions per unit conditioned roof area. The annual
heating and cooling energy uses of four commercial
building prototypes—new office (1980+), old office
(pre-1980), new retail (1980+), and old retail (pre-
1980)—were simulated in 236 US cities. Substituting
a weathered cool white roof (solar reflectance 0.55)
for a weathered conventional gray roof (solar reflec-
tance 0.20) yielded annually a cooling energy saving
per unit conditioned roof area ranging from
3.30 kWh/m2 in Alaska to 7.69 kWh/m2 in Arizona
(5.02 kWh/m2 nationwide); a heating energy penalty
R. Levinson (*) : H. Akbari
Lawrence Berkeley National Laboratory,
1 Cyclotron Road, MS 90R2000,
Berkeley, CA 94720, USA
e-mail: RML27@cornell.edu
ranging from 0.003 therm/m2 in Hawaii to 0.14 therm/
m2 in Wyoming (0.065 therm/m2 nationwide); and an
energy cost saving ranging from $0.126/m2 in West
Virginia to $1.14/m2 in Arizona ($0.356/m2 nation-
wide). It also offered annually a CO2 reduction ranging
from 1.07 kg/m2 in Alaska to 4.97 kg/m2 in Hawaii
(3.02 kg/m2 nationwide); an NOx reduction ranging
from 1.70 g/m2 in New York to 11.7 g/m2 in Hawaii
(4.81 g/m2 nationwide); an SO2 reduction ranging
from 1.79 g/m2 in California to 26.1 g/m2 in Alabama
(12.4 g/m2 nationwide); and an Hg reduction ranging
from 1.08 μg/m2 in Alaska to 105 μg/m2 in Alabama
(61.2 μg/m2 nationwide). Retrofitting 80% of the 2.58
billion square meters of commercial building condi-
tioned roof area in the USA would yield an annual
cooling energy saving of 10.4 TWh; an annual heating
energy penalty of 133 million therms; and an annual
energy cost saving of $735 million. It would also offer
an annual CO2 reduction of 6.23 Mt, offsetting the
annual CO2 emissions of 1.20 million typical cars or
25.4 typical peak power plants; an annual NOx
reduction of 9.93 kt, offsetting the annual NOx
emissions of 0.57 million cars or 65.7 peak power
plants; an annual SO2 reduction of 25.6 kt, offsetting
the annual SO2 emissions of 815 peak power plants;
and an annual Hg reduction of 126 kg.
Keywords Air pollution . Building energy simulation .
Carbon dioxide . Cool roof . Cooling energy . Energy
cost . Greenhouse gas . Heating energy . Mercury .
Nitrogen oxides . Solar reflectance . Sulfur dioxide
54
Introduction
Roofs that have high solar reflectance (high ability to
reflect sunlight, spectrum 0.3–2.5 μm) and high thermal
emittance (high ability to emit thermal radiation,
spectrum 4–80 μm) stay cool in the sun. Low roof
temperatures lessen the flow of heat from the roof into
the building, reducing the need for electricity for space
cooling in conditioned buildings. Cool roofs may also
increase the need for heating energy in cold climates.
Since building heat gain through the roof peaks in mid-
to-late afternoon, when summer electricity use is high-
est, cool roofs can reduce peak electricity demand.
Measurements in several warm weather climates,
including California, Florida, and Texas, typically
yielded summertime daily air conditioning savings and
peak demand reductions of 10% to 30%, though values
have been as low as 2% and as high as 40% (Konopacki
et al. 1998). Measured energy savings and peak
demand reductions are summarized by Levinson et
al. (2005).
The extents to which replacing a conventional
(hot) roof with a cool roof will reduce the need for
cooling energy and/or increase the need for heating
energy depend on building construction, building
operation, and climate. Prior research has indicated
that net annual energy cost savings are greatest for
buildings located in climates with long cooling
seasons and short heating seasons, particularly those
buildings that have distribution ducts in the plenum
(Akbari et al. 1998, 1999; Konopacki and Akbari
1998).
The combustion of fossil fuels in a power plant or
building furnace produces greenhouse gases and air
pollutants, including but not limited to carbon dioxide
(CO2), nitrogen oxides (NOx), sulfur dioxide (SO2),
and mercury (Hg). Emission reductions achieved
through the use of cool roofs vary with both energy
saved and the local mix of fuels used to generate
electricity. For example, conserving equal amounts of
cooling electricity reduces Hg emissions more in a
region served by coal-fired power plants than in an
area with natural gas power generation. The net
annual energy cost saving also depends on regionally
varying energy prices.
The influence of cool roof installation on a
building’s energy use is proportional to its “condi-
tioned roof area” (CRA), or area of roof over
conditioned space. While cooling energy savings
Energy Efficiency (2010) 3:53–109
and heating energy penalties per unit CRA can be
estimated through building energy simulations, local
rates of energy savings and penalties per unit land
area (LA) depend on building density, or ratio of
conditioned roof area to land area. Regional sums,
such as state cooling energy savings, are the land area
integrals of these per-LA rates.
This study combines building energy simulations,
local energy prices, local electricity emission factors
(mass of greenhouse gas or air pollutant emitted per
unit energy supplied to the grid), and local estimates
of building density to characterize local per-CRA and
per-LA annual rates of cooling energy saving, heating
energy penalty, energy cost saving, and emission
reduction in the USA. We also report regional CRAs
and regional average per-CRA rates—ratios of re-
gional saving or penalty (land area integral of per-LA
rate) to regional CRA—for each US state and for the
nation.
Theory
Cooling and heating load changes per unit
conditioned roof area
Consider a building with electric cooling and natural
gas (hereafter, simply “gas”) heating. Installing a cool
roof will reduce the building’s solar heat gain,
decreasing its annual cooling load (amount of heat
that must be removed by its cooling equipment) while
increasing its annual heating load (amount of heat that
must be supplied by its furnace).
The influence of a cool roof on the building’s
conditioning energy use is quantified in this study by
comparing simulations of the building’s cooling energy
and heating energy uses with a cool roof to those with a
conventional roof. Replacing a conventional roof by a
more solar-reflective cool roof reduces roof solar heat
gain by Δr
A
I, where Δr is the increase in roof
solar reflectance, A is the roof area, and I is the solar
irradiance. We expect that the local reduction in annual
cooling load will be roughly proportional to Δr
A

R1
I ðx; yÞ
C ðx; yÞ, where R is the thermal resis-
tance of the roof assembly; I ðx; yÞ is the local annual
mean global horizontal solar irradiance (Fig. 1);
C ðx; yÞ is the local number of annual cooling degree
days (Fig. 2a); and ðx; yÞ locates the building in a
projected (flat) Earth coordinate system. Similarly,
Energy Efficiency (2010) 3:53–109 55

146
130

2
8
6 158
15

15
13
140
164
2 154

156

160
14

15
0 152

4
14

15

170
166
148

164
158 156

154
16

2
WA

15
17

164

4
2 160 ME

16
2 168 156
16 MT
180 ND 15
2

154
166
178

16 156 4
OR 162 0 158 16
172 156

162
VT NH
182 ID MN 156
176
18
17

WI 158 154 MA 164

0

158
NY
184
0

SD

158
162
8 162

154
16 MI RI
17 156 CT
17 4

164
17
19
2
194
WY 8 160 156

156
0

16
182
160
19

160

4
21

154
164
2

0 202 IA 2 PA NJ
16

8
200 NE

16
16
10 NV 20 OH

2
0 8 2 4 MD 170DE
20 20 16

196

172
UT IN

158
IL 4
18 WV

198
CA CO 8 VA

182

174

182
208
8 KS MO

18
22 21
8

4
222

KY

20

6
21
232
23

6
216 180
234

17
216

4
18
2
24

212 230 216

176
21 21

184
NC
218
20
8 4

23 TN
8

180
228 226

2
OK
214
24 230

19
AZ 0 SC
8

232 NM AR
23

19
0 192

2
23 19

19
6
6
212
234

224
240

19 MS AL 192 GA
23

2
8
8

23

6
23

222
0
24

2
8 19
18 18

19
TX 6 4

196
19
192 LA
220

190
196

20
188
206 200

2
6
21
4 20 190
0

204
20

200
8 194 FL
19

198
192
4

202
10

204
8
AK 20
198
102

10
106

8 226
0
90

11

annual mean global horizontal solar irradiance

96

104
2
228 W/m
98
2
10

High : 244
202
200

194
114

HI
Low : 90

Fig. 1 Annual mean global horizontal solar irradiance (W/m2) computed from NREL TMY2 typical meteorological year data (NREL
2007)

we expect the local increase in annual heating Thus, cool roof energy savings and penalties are
load to be roughly proportional to Δr
A
R1
proportional to roof area rather than to floor area.
I ðx; yÞ
H ðx; yÞ, where H ðx; yÞ is the local number Since the thermal conductivity of roof insulation
of annual heating degree days (Fig. 2b). typically rises with temperature, a cool roof surface
Cool roof changes to a building’s annual cooling can lower the temperature and thus increase the
and heating loads depend on the thermal resistance thermal resistance of roof insulation (Levinson et al.
and thermal capacity of the roof assembly; the 1996). The DOE-2.1E building energy simulations
operating schedules of the building and its HVAC performed for this study incorporate a roof assembly
system; the efficiencies of the HVAC equipment; and heat transfer module that accounts for this effect
the climate. A cool roof can decrease annual cooling (Gartland et al. 1996).
load even in climates with zero nominal (base 18°C) The ratio of annual heating load increase to annual
cooling degree days if air conditioning is required to cooling load decrease is a simple measure of the
remove the building’s solar heat gain and/or internally influence of cool roof installation on a building’s
generated heat. Note also that since there is negligible annual heat balance. If the installation of a cool roof
heat transfer between floors in a conditioned building, reduces annual cooling energy (electricity) use per
roof solar heat gain affects the heat balance of only CRA by eðx; yÞ [kWh/area; 1 kWh=3.6 MJ] and
the top floor in a multi-storey conditioned building. increases annual heating energy (gas) use per CRA by
56 Energy Efficiency (2010) 3:53–109

100

0
50

15

0
10

50

100
WA
50
150 150

200
20
0

200
150
35

0
15

300
0

15
0

150
ME

0
MT

25
ND 100
OR 150 50

0
300

20
30

15
200 200

30

0
50

0
0
100 35 MN 15 250 VT NH
0 0 0
30

20
350

0
ID

0
25
250
25

250
WI 250 MA

300
NY
0

300
50

SD

25
20
WY MI

200
0 0 CT RI

0
10 20 0
30

40
200 450

0
200

250
550

15
0
0
35
250

55 300

15
0
250

20

550
0
50

0 IA NJ

0
400

35
20 PA
0 NE
550

450
35

10 0

CO 0
15

OH DE
0

NV

20 0
5 30

60
60 1 0 600
0

500

0
UT MD

0
30

5
IN
0
0

400

40
350

IL
50
55

450
300
55

0
1,0

0
400

CA 0

55
60
00 1,20

400 450

0
200 1,4 VA

70
45 WV
50 1, 20 KS
75

0 50 MO
0

150

1,0 50 0 70
35
0

1,
650

1,350

0
100

40 KY 800

55
250
0
500

0
0

50

30
1,45

NC
0

500 0
400

80

0
0
1,15 0

500
55
250

AZ 1,000 TN

1,000
80
550

0 900
1,050 950 OK

1,
1,

10
450

1,9

1,2 AR SC
50

10 NM

0
0 850 00
1,6

50
1,2

1,4

1, GA

00
1,300

10
00 1

1,250
00

1,1
1,3
0
1,150

50
0
15

MS
,45

1,350
1,200

AL
1,

50
1,100
0

1,1
0
1,25
0

0
35

1,40
TX

00
35
1,500
1,

1,450 LA 0
1,50

1,

1,3
1, 300 00

1,40
4
1, 1,60
0
0

0
,30 1,7
50

1 50
1,5

0 1,6 FL 0
50
60 5
1, 00 1,8 0
1,8 10
2,
00

0
25
1,7

00 2,
1,8 50
AK 1,9

cooling degree days

CDD18C
High : 2,640
00
1,8
HI
Low : 0

Fig. 2 Maps of a annual cooling degree days (CDD18C) and b annual heating degree days (HDD18C) computed from NREL TMY2
typical meteorological year data (NREL 2007)

gðx; yÞ [therm/area; 1 therm=105.5 MJ], the dimen- solar reflectance of a roof’s surface from r0 ¼ 0:20
sionless “load change” ratio is to r1 ¼ 0:55 will reduce the roof’s solar heat gain by
ðr1  r0 Þ=ð1  r0 Þ ¼ 44%, but will decrease the
cooling energy use and increase the heating energy
h1
h g ðx; yÞ
‘ðx; yÞ ¼ ; ð1Þ use of a single-storey office or retail building by only
eðx; yÞ
EER
0:01 therm=kBTU about 5–10%. Fractional cool roof cooling energy
savings and heating energy penalties are even
where hh is the dimensionless efficiency of the smaller for multi-storey buildings. However, the cost
heating equipment and EER (energy efficiency ratio) of a roof reflectance upgrade is also only a small
is the dimensional coefficient of performance of the fraction of the total cost of a building. It is simplest
cooling equipment [BTU/(h·W); 1 BTU=1.055 kJ]. to evaluate the economics of a cool roof upgrade by
comparing the present value of the building’s lifetime
Fractional cooling energy savings and heating energy energy cost savings to the cost of the upgrade.
penalties
Energy cost saving per unit conditioned roof area
Any single building component is typically respon-
sible for only a small fraction of the building’s total The installation of a cool roof yields a per-CRA
energy consumption. For example, increasing the annual energy cost saving (monetary value of cooling
Energy Efficiency (2010) 3:53–109 57

00 00
b
2
3,

2,8

5,200
4,600
00

4,200
3,0 WA

5,400
00
2,6 4,400

3,800

5,0

5,200
4,200
00

3,000
5,2
MT 0 00

4,6
4,40 5,600 ME
3,6

4,8
00
5,0 00
00

00
4,6

4,000
ND 00

00
00

VT
2,6

3,8

4 ,2
MN
OR 00 4,200

4,6
3,600

3,4 4,8 NH
00 00 4,00

00
0
0 ID MA 3,400
,60 4,200
2,8

0
3,200

3 WI 4,00 0
SD ,60

3,400
0
00

4,000 3,80
0 3

00
WY RI

4,
MI 0 0 NY CT

00
8
4,4 3,

4,20
00

3,8
4,00
3,4 3,600 PA
3,200

0
4, IA NJ

4,0
00
0 000 3,600
1,600

20 0
3,40

00
4, 4,2 NE
NV 00 2,800
3,

3,40
0
OH MD
00

4,40
0

3,400
UT IN
40
0

IL DE

0
3,

CA CO 3,200 WV
3,6

4,20 VA

2,0
1, 0
00

60 KS MO
0

1,40

00
4,40

0 0 KY
0 0 2,600
1,000 3,
40 4,0
NC
00

2,4
0
2,800 2,40
3,6

3,200 TN

00
80
0 2,20 OK
1,200

AZ 80 0
00

1,600
0 NM AR SC
1,800
1,0

1,600
1,0 2,000
1,200

00
AL GA
1,600

MS

1,200
00
,4
1,600

1 0
00
TX 1,00 1,

00
0 LA 0

1,0
0
20 FL 80
1, 0
0 60
0
1,4

0
40
0
7,60

40
0
80
0

200
7, AK
40
0

0
7,20 7,6
60

0
7,0
00 400
7,4

0
6,600 0
00
6,800

0
20

6,20
7,

heating degree days

6,400
HDD18C
High : 7,890

HI
Low : 0

Fig. 2 (continued)

energy saving less monetary value of annual heating lifetime energy cost saving is computed by treating
energy penalty) of each year’s cost savings as an ordinary annuity. Given
a real (inflation-adjusted) annual rate of return r, the
cðx; yÞ ¼ eðx; yÞ de ðx; yÞ  gðx; yÞ dg ðx; yÞ ð2Þ present value (PV) of N years of constant annual
energy cost saving cðx; yÞ is b
cðx; yÞ, where
where de ðx; yÞ and dg ðx; yÞ are the annual average
X
N h i.
prices of electricity and gas, respectively. This b¼ ð1 þ rÞi ¼ 1  ð1 þ rÞN r: ð3Þ
relationship is approximate, since energy prices vary i¼1
over the course of year.
The PV multiplier b increases with roof lifetime N
Life cycle cost saving per unit conditioned roof area and decreases with real annual rate of return r
(Table 1).
The life cycle cost saving of a cool roof is equal to the
present value of its lifetime energy cost savings minus Emission reductions per unit conditioned roof area
any initial cost premium. The cost premium for
choosing a cool white roofing product rather than a If a power plant emits greenhouse gas or air-pollutant
conventional roofing product is typically 0 to $2.20/ i at an annual average rate of fe;i ðx; yÞ mass units per
m2 (Levinson et al. 2005). The present value of the unit electrical energy supplied to the grid and a gas
58 Energy Efficiency (2010) 3:53–109

Table 1 Present value multiplier b (ratio of present value of savings or heating energy penalties of the prototypes
lifetime energy cost saving to annual energy cost saving)
k ¼ 1 . . . n. That is,
computed from Eq. 3 for various combinations of roof lifetime
N and real (inflation-adjusted) annual rate of return r X
n
eclass;j ðx; yÞ ¼ vj;k eprototype;k ðx; yÞ ð7Þ
Roof lifetime N (years) Real annual rate of return r
k¼1
3% 5% 7%
and
15 11.9 10.4 9.1 X
n
20 14.9 12.5 10.6 gclass;j ðx; yÞ ¼ wj;k gprototype;k ðx; yÞ ð8Þ
25 17.4 14.1 11.7 k¼1
30 19.6 15.4 12.4
where vj;k and wj;k are location-independent coeffi-
cients relating the cooling energy savings and heating
energy penalties (respectively) of the building classes
furnace emits the same substance at a rate of fg;i mass to those of the prototypes. In matrix form,
units per unit gas energy consumed, the installation of
Eclass ðx; yÞ ¼ V Eprototype ðx; yÞ ð9Þ
a cool roof will reduce the annual per-CRA mass of
substance i emitted by and

pi ðx; yÞ ¼ eðx; yÞ h1

t fe;i ðx; yÞ  g ðx; yÞ fg;i ; ð4Þ Gclass ðx; yÞ ¼ W Gprototype ðx; yÞ ð10Þ
where
where ht is the electrical grid’s transmission efficiency 2 3
(ratio of output to input). eclass;1 ðx; yÞ
6 .. 7
Eclass ðx; yÞ  4 . 5;
Extrapolation of energy savings and penalties eclass;m ðx; yÞ
2 3 ð11Þ
The influence of a cool roof on energy use depends eprototype;1 ðx; yÞ
not only on the building’s location (a proxy for 6 .. 7
Eprototype ðx; yÞ  4 . 5;
climate) but also on its construction and operation.
Energy use simulations are typically conducted for a eprototype;n ðx; yÞ
limited number n of building prototypes, often far
fewer than the m classes of buildings present in some 2 3
region of interest. gclass;1 ðx; yÞ
6 .. 7
If two co-located buildings “1” and “2” differ only Gclass ðx; yÞ  4 . 5;
in weekly occupancy time t, roof assembly thermal gclass;m ðx; yÞ
resistance R, cooling energy efficiency ratio EER,
2 3 ð12Þ
and/or heating efficiency hh, we expect that gprototype;1 ðx; yÞ
6 .. 7
e1 ðx; yÞ t1 EER2 R2 Gprototype ðx; yÞ  4 . 5;
ffi

ð5Þ gprototype;n ðx; yÞ
e2 ðx; yÞ t2 EER1 R1
and
and 2 3 2 3
v1;1   v1;n w1;1   w1;n
g1 ðx; yÞ t1 hh;2 R2 6 7 6 7
ffi

: ð6Þ V  4 ... . . . ... 5; W  4 ... . . . ... 5: ð13Þ
g2 ðx; yÞ t2 hh;1 R1
vm;1   vm;n wm;1   wm;n

We assume that the per-CRA annual cooling If the energetics of building class j are related to
energy saving or heating energy penalty of each the energetics of a single prototype k, the matrices V
building class j can be expressed as a linear and W will be sparse, with only one nonzero entry per
combination of the per-CRA annual cooling energy row.
Energy Efficiency (2010) 3:53–109 59

Rates per unit land area census division containing point ðx; yÞ, we assume
that
The rates of cooling energy saving, heating energy
penalty, energy cost saving, and emission reduction U ðx; yÞ ¼ hðx; yÞ Rðx; yÞ ð15Þ
per unit LA depend on the density and type
distribution of the local building stock. Let matrix where
2 3 2 3
u1 ðx; yÞ r1 ðx; yÞ
6 .. 7 6 .. 7
U ðx; yÞ  4 . 5 ð14Þ Rðx; yÞ  4 . 5: ð16Þ
um ðx; yÞ rm ðx; yÞ
where uj ðx; yÞ is the ratio of CRA to LA for
building class j. Since building inventory statistics The per-LA rates of annual cooling energy saving
typically characterize broad areas, such as US and heating energy penalty are then
census divisions (Fig. 3), we use county-level
population density hðx; yÞ (Fig. 4) to better estimate e^ðx; yÞ ¼ U ðx; yÞT Eclass ðx; yÞ
local ratios of CRA to LA. That is, if rj ðx; yÞ is the h i
¼ U ðx; yÞT V Eprototype ðx; yÞ ð17Þ
ratio of CRA of building class j to population in the

Pacific New England

(9) (1)

West North Central

(4) Middle Atlantic
(2)
Mountain
(8) East North Central
(3)

East South Central South Atlantic

(6) (5)

West South Central

(7)

Pacific
(9)

U.S. census divisions

Pacific
(9)

Fig. 3 Boundaries of the nine US census divisions (ESRI 2007b)

60 Energy Efficiency (2010) 3:53–109

10 50

2
10 WA
100

10
50 20 2

1
5 2
200 MT 5 2

2
5
20

25
2

5
10
10 100

2
10
1 10 2 5 ME

10
10
10
20

50 2

2
1 10

10 5
10 ND 5

10
5
5 10
2
2 5

20
1

10
10
5

5
10 10

20
20 20

20
0
10

5
10 1

5
1

10 5

10
ID

5
5 20 VT NH
20

5
OR 10

20
2

10
5
50
200

10
500 5

5
1 10 MN WI 10 50
20

20

2 200
20 50
50
20

20
52 100

5
SD

20

500
10 10 MA

50

100
50
2
10

50
0

20
1 20

1
10 20

10
10 5

1
50

200
10

0
5
50 50

50
WY 5 10

50
20
20 10

50
100
10

RI
2
10
2

5 50 200

50
2

200
NY

20
20
10

5
10
MI 1,000 20 20 10 CT

100
2

50 10
20 1

0
10
0 0

10
1
20 IA PA

100
2,0 50 00 500

20 50
50

5
1
5 0

1
2

20
10 0 0 0 1

200
1
1 2
2

1
5 200

10
1

1
100

50
20
100

200
100 MD 100 NJ
100

50
1
200

20
5 20 100 50

50 100
20 IN 0
10

20 20 20

50
2 10 5 20
500

20
1 50 1 55 10 50 DE
50 50 100 10 2 0

100
NV 10 5

500
20

50
50

20
20
1 1 5 10 10

50
50

200
200

UT 10 10 NE 50 20 WV 00
200 10

100
500

50
5
5 10 5 20

20
5

20 IL 50 OH

10

50 20
100
1 2 20

2
20

100
10

5
CO 50

10
50

200
10

20
100
0

200
100 50

50
10
10 20
5 100 50

10

20 50
50

20
50 10

50
CA

50
10
5

2 KS 10

20
10

10
10 20 VA 20

50
5 50 10

2
5

20
20
1

10
5
5

50
10 20 2 50 20 100

50
50 50 10 0 5

5
10 KY

10
10

5
500

5
1 10 10 100 0

50
5

5 MO 10 0
5

1
NC
20 10 50
5

20
2
10 10

2
100

50

50
50
50 0

500
10 TN 1 20

100

20

20 20
100
1,000

50

20
20
10 20 5 20 200 50
20

200
20

200
1
1 5

20
20

100

50
50

20
1
10 1

10
50 20

50
AZ 20

50
50
OK 20

10
50
2 0 1 0 1
200

10
0

50
5 5 50 50 00

0
0
100

NM

50
2

10
5

SC 50
5 10

50

20 10
20 10 5 0

0
AR 50 10 2 10
50

50
10

10
10
20 20

20
20
100 0

20
2

AL

5
0

10 10 0
50 1 20 10
12 2 50
51 0

10
50
20

20
10

10
10 5
5

5
20 2
50
50 100 10
1

100
1

50

50
50

50
50
5

10
2 1

50 5 20
50 20 20
100

10 50
5
50

5
5 1 5 10 0 GA
20 5
10

50
50

20
MS 20

20
LA

50
TX 20 100 20 10 50 20 20 0

100
50 10 5
1

10
10

50

5
5 1 20
50
100

10
100
50
0 10

10 50
100
20

50
2

0
50

50
5
20 5
200
10

200

10
10
0

0
20

50
0

2
10

FL 20
5

20 20 5 200
1
5 10

50 200 5
500
1
5 10
2

50
50
AK 2
1 1
2
1

population per unit land area

1

2
persons / km
1
1 High : 21,500
1

HI
Low : 0

Fig. 4 US population density (persons/km2) from the 2004 US Census (ESRI 2007b)

and aggregate annual cooling energy saving, heating

energy penalty, energy cost saving, and emission
g^ ðx; yÞ ¼ U ðx; yÞT Gclass ðx; yÞ reduction. Dividing each regional sum by aggregate
h i regional CRA yields the regional-average per-CRA
¼ U ðx; yÞT W Gprototype ðx; yÞ ð18Þ
rate.

respectively. These yield per-LA annual rates of

energy cost saving and emission reduction Methodology
c^ðx; yÞ ¼ e^ðx; yÞ de ðx; yÞ  g^ ðx; yÞ dg ðx; yÞ ð19Þ
Estimating prototype energy saving and heating
and penalty per unit conditioned roof area
p^i ðx; yÞ ¼ e^ðx; yÞ h1
t fe;i ðx; yÞ  g ðx; yÞ fg;i
^ ð20Þ
In a year 2005 study (Akbari and Konopacki 2005),
respectively. one of the authors used the DOE-2.1E building
The local per-LA annual rates e^ðx; yÞ, g^ ðx; yÞ, energy model (DOE-2 2007) to simulate for a typical
^cðx; yÞ, and p^i ðx; yÞ can be integrated over any region meteorological year (NREL 2007) the hourly heating
of interest, such as a US state, to obtain the region’s and cooling energy uses of four prototype single-
Energy Efficiency (2010) 3:53–109
storey commercial buildings, including both new
(1980+) and old (pre-1980) offices and retail stores.
The general characteristics of the office and retail
store prototypes were based on survey data from the
California Energy Commission (CEC 1994). Loca-
tion-specific properties such as roof and wall con-
struction, window characteristics, and building
schedules were obtained from the US Energy Infor-
mation Administration (EIA 1979, 1983, 1994a).
Additional office and retail store attributes were
derived from building characteristic surveys con-
ducted in both northern and southern California
(Akbari et al. 1989, 1991, 1993).
Each building was simulated with several different
levels of roof insulation thermal resistance. The
prototypes used in the current study—new office,
old office, new retail, and old retail—are identical to
the prototypes described in the prior study, except that
in the current study, the roof insulations in the new
and old buildings are assigned thermal resistances of
3.3 m2 K W−1 [R-19] and 1.2 m2 K W−1 [R-7],
respectively. The prototypes are partly characterized
in Table 2 and fully detailed in the year 2005 study.
Akbari and Konopacki (2005) simulated the annual
heating and cooling energy uses of each prototype
twice: first, with a weathered conventional gray roof
(solar reflectance 0.20, thermal emittance 0.90) and
then with a weathered cool white roof (solar reflec-
tance 0.60, thermal emittance 0.90). Prototype energy
uses were evaluated in each of 236 US cities (Fig. 5)
to yield per-CRA cooling energy saving (annual
Table 2 Major characteristics of the prototype buildings whose annual cooling and heating energy uses were simulated with cool and
conventional roofs
Geometry
Roof area and floor area (m2)
Roof construction
Thermal resistance of roof insulationa (m2 K/W)
Thermal resistance of wall insulation (m2 K/W)
Cooling equipment
Cooling energy efficiency ratio (BTU/[h·W])
Heating equipment
Heating efficiency (%)
Operating hours
a
Roof insulation levels differ from those of prototypes described by Akbari and Konopacki (2005)
61
cooling energy use with a conventional roof minus
annual cooling energy use with a cool roof) and per-
CRA heating energy penalty (annual heating energy
use with a cool roof minus annual heating energy use
with a conventional roof) when roof solar reflectance
is increased by 0.40 (to 0.60 from 0.20). The
simulations are fully detailed in the year 2005 study.
While it is reasonable to assign a solar reflectance of
0.60 to a weathered white roof, lowering this value to
0.55 makes estimates of cool roof energy saving more
conservative and consistent with the nonresidential
cool roof analysis we performed for California’s Title
24 building energy efficiency standards (Levinson et
al. 2005). A building’s cooling energy saving and
heating energy penalty are each proportional to the
increase in the solar reflectance of its roof (Konopacki
et al. 1997). Hence, we scaled these results by a factor
of 0.35/0.40=0.875 to estimate the per-CRA cooling
energy saving and heating energy penalty when roof
solar reflectance is raised by 0.35 (to 0.55 from 0.20).
All savings, penalties, and emission reductions in this
study are based on increasing the weathered solar
reflectance of a roof to 0.55 from 0.20.
The geographic information system (GIS) applica-
tion ESRI ArcGIS Desktop 9.1 (ESRI 2007a) was
used to create spatial maps of per-CRA annual
cooling energy saving ek ðx; yÞ and heating energy
penalty gk ðx; yÞ for each prototype k. Each map is a
raster of 5×5-km cells spanning the USA; cell values
were populated by inverse-distance-weighted interpo-
lation between values simulated in each of the 236
New office Old office New retail store Old retail store
(1980+) (pre-1980) (1980+) (pre-1980)
Single-storey, non-directional, Single-storey, non-directional, one zone
five zones (conditioned)
455 753
Built-up materials on flat deck
3.3 [R-19] 1.2 [R-7] 3.3 [R-19] 1.2 [R-7]
2.3 [R-13] 1.1 [R-6] 2.3 [R-13] 0.7 [R-4]
Package a/c, direct expansion, air cooled
10 8 10 8
Gas furnace
74 70 74 70
Weekdays 6A-7P Weekdays 8A-9P, weekends 10A-5P
62 Energy Efficiency (2010) 3:53–109

Quillayute

Seattle Caribou
Olympia Kalispell Cut Bank
Spokane
Astoria
Yakima
WA
Glasgow International Falls
Minot
Portland Missoula Great Falls ME
Salem Pendleton Helena Lewistown
MT ND Houghton
Eugene Fargo Duluth Sault Ste. Marie
North Bend Redmond Miles City Bismarck Massena
Billings Portland
OR Burns
MN BurlingtonVT NH
Alpena
Medford BoiseID Sheridan
Traverse City
Saint Cloud Eau Claire Boston
Green Bay SyracuseNY AlbanyMA
Arcata Rapid CitySD Pierre Huron Minneapolis Providence
Pocatello RochesterWI MI
Buffalo Rochester Hartford
Lander Casper Sioux Falls
Muskegon
Flint Binghamton CT RI
Mason CityLa Crosse Madison
Winnemucca WY Detroit Erie Bradford
Bridgeport
Waterloo Rockford Milwaukee Lansing Newark
Elko Rock Springs Sioux City Youngstown New York City
Scottsbluff Chicago Toledo
Reno Salt Lake City Norfolk
IA PA NJ
Cheyenne Omaha Moline Mansfield Akron
Sacramento NE North Platte Des Moines Pittsburgh Atlantic City
Wilmington
Ely Peoria Fort Wayne OH Columbus
San Francisco NV Boulder
Grand Island
MD DE
UT Eagle
IL SpringfieldIN
Tonopah Dayton Elkins
Grand Junction DC
Goodland Covington
CA Fresno WV
Cedar City CO Colorado Springs Topeka Columbia St. Louis Richmond
Pueblo
Evansville
LouisvilleLexington Roanoke VA Norfolk
KS MO Lynchburg
Alamosa Dodge City KY
Santa Maria Bakersfield Las Vegas Wichita
Springfield Bristol GreensboroRaleigh Cape Hatteras
Daggett
Nashville Knoxville NC
Los AngelesLong Beach Flagstaff Tulsa
Prescott TN Charlotte Wilmington
Albuquerque Tucumcari Amarillo Oklahoma City Fort Smith ChattanoogaAsheville Greenville
OK Memphis
San Diego AZ Little Rock
Huntsville
AthensSC
Columbia
Phoenix NM AR
Atlanta Augusta Charleston
Lubbock Wichita Falls Birmingham
Macon
MS MeridianAL
Tucson Savannah
Fort Worth GA
El Paso Abilene Shreveport
Midland Montgomery
Jackson
San Angelo Waco Jacksonville
TX Lufkin
Mobile Tallahassee
LA Baton Rouge
Austin Daytona Beach
HoustonLake Charles New Orleans
San Antonio Port Arthur
FL
Victoria Tampa
Nome West Palm Beach
Corpus Christi
Miami
St. Paul Island Bethel Lihue
McGrath Fairbanks Brownsville
Key West
AK Talkeetna Big Delta
Cold Bay Anchorage Gulkana Honolulu

Kodiak

Yakutat Kahului
simulation cities

HI Hilo
Annette

Fig. 5 Locations of the 236 US cities in which building energy use was simulated with conventional and cool roofs

cities. Each city was placed in the x–y plane via NAD the District of Columbia, which we treat as a state)
1983 Lambert Conformal Conic projection of its were obtained from the Energy Information Admin-
latitude and longitude. istration (EIA 2006a, 2007a). Raster maps of elec-
All raster maps produced in this study share the tricity price de ðx; yÞ and gas price dg ðx; yÞ were
same domain, cell size, and projection. generated by assigning the appropriate EIA price to
each state’s raster cells (Fig. 6).
Estimating prototype load change ratios A raster map of per-CRA annual energy cost
saving ck ðx; yÞ was created for each prototype k via
Equation 1 was used to compute the load change ratio Eq. 2.
‘k ðx; yÞ (dimensionless ratio of annual heating load
increase to annual cooling load decrease) of each Estimating prototype emission reduction per unit
prototype k. conditioned roof area

Estimating prototype energy cost saving per unit Table 3 presents year 2004 electricity emission factors
conditioned roof area (mass of greenhouse gas or air pollutant emitted per
unit electrical energy supplied to the grid) for CO2,
Year 2005 average commercial sector prices of NOx, SO2, and Hg obtained from eGRID2006v2.1,
electricity and natural gas in each US state (and in the latest version of the Emissions & Generation
Energy Efficiency (2010) 3:53–109 63

Table 3 Electricity emission factors (mass of greenhouse gas or air pollutant emitted per unit energy supplied to the grid) and
generation resource mixes for each subregion of the US EPA eGRID dataset

Subregion Emission factor Generation resource mix

CO2 NOx SO2 Hg Coal Oil Gas Other Nuclear Hydro Non-hydro Unknown
(kg/kWh) (g/kWh) (g/kWh) (μg/kWh) (%) (%) (%) fossil (%) (%) (%) renewable (%) (%)

AKGD 0.570 1.35 0.60 0.8 12.3 7.3 68.0 0.0 0.0 12.4 0.0 0.0
AKMS 0.218 2.96 0.30 0.0 0.0 28.8 3.6 0.0 0.0 66.9 0.7 0.0
AZNM 0.569 0.94 0.65 11.5 40.4 0.0 31.5 0.0 21.2 4.5 2.4 0.0
CAMX 0.399 0.34 0.25 1.0 12.6 1.1 46.4 0.9 14.2 15.1 9.7 0.1
ERCT 0.644 0.44 1.43 13.2 37.7 0.5 45.9 1.3 13.2 0.3 1.0 0.2
FRCC 0.602 1.02 1.64 4.1 26.4 18.3 36.5 0.3 15.5 0.0 2.0 1.0
HIMS 0.661 3.17 2.71 0.0 3.6 77.2 4.1 0.0 0.0 3.0 12.1 0.0
HIOA 0.784 1.16 1.59 7.3 18.0 77.4 0.0 1.9 0.0 0.0 2.7 0.0
MROE 0.843 1.47 3.40 13.9 71.3 2.4 5.2 0.1 13.2 3.9 3.9 0.1
MROW 0.823 1.73 2.66 19.6 74.6 0.6 1.8 0.1 16.0 4.7 2.1 0.0
NEWE 0.412 0.44 1.06 3.9 14.5 9.4 36.7 1.0 27.6 5.1 5.7 0.0
NWPP 0.418 0.73 0.57 4.4 34.4 0.3 10.6 0.1 3.6 49.0 2.0 0.0
NYCW 0.418 0.40 0.31 2.9 0.0 20.4 29.8 0.3 48.6 0.0 0.8 0.0
NYLI 0.641 0.83 2.43 2.6 0.0 58.2 35.5 1.8 0.0 0.0 4.5 0.0
NYUP 0.372 0.45 1.90 6.4 25.4 6.6 13.2 0.3 27.1 26.0 1.4 0.0
RFCE 0.497 0.77 3.64 18.7 44.9 3.5 9.6 0.7 38.4 1.6 1.4 0.0
RFCM 0.745 1.11 3.09 14.9 67.0 0.9 15.5 0.3 14.3 0.0 2.0 0.0
RFCW 0.706 1.28 4.62 19.8 72.8 0.5 1.5 0.7 23.2 0.7 0.4 0.1
RMPA 0.923 1.41 0.92 7.4 80.6 0.0 13.5 0.0 0.0 5.3 0.5 0.0
SPNO 0.894 1.80 2.74 12.4 78.1 1.3 4.6 0.1 15.2 0.1 0.5 0.0
SPSO 0.799 1.15 1.79 17.6 58.8 0.2 34.1 0.3 0.0 4.2 2.3 0.1
SRMV 0.515 0.66 1.03 5.0 23.4 5.0 39.3 1.1 26.6 1.6 2.4 0.5
SRMW 0.837 1.14 3.15 18.5 84.7 0.3 2.0 0.1 11.7 1.2 0.1 0.0
SRSO 0.676 0.98 3.83 15.8 64.0 0.6 10.1 0.1 18.6 3.1 3.5 0.0
SRTV 0.678 1.17 3.27 11.5 65.8 1.7 2.4 0.0 20.4 8.8 0.9 0.0
SRVC 0.520 0.84 2.66 9.9 51.0 1.7 3.8 0.2 39.5 1.7 2.0 0.1

Resource Grid Integrated Database released by the A map of per-CRA emission reduction pi;k ðx; yÞ
US Environmental Protection Agency (EPA 2007). was created for each combination of substance i and
eGRID assigns the land within each US ZIP code to prototype k using Eq. 4.
one of 26 subregions (Fig. 7). A raster map of
electricity emission factor fe;i ðx; yÞ for each substance Estimating conditioned roof area per unit land area
i was created by assigning the subregion’s emission
factor to the raster cells within that subregion The 2003 Commercial Buildings Energy Consump-
(Fig. 8). tion Survey (CBECS) public use microdata files
Table 4 shows the non-regional natural gas
emission factors fg;i (mass of greenhouse gas or air Table 4 Non-regional natural gas combustion emission factors
(mass of greenhouse gas or air pollutant emitted per unit energy
pollutant emitted per unit gas energy consumed)
consumed)
obtained from the US EPA (EPA 2005). We chose
an NOx factor in the middle of the range of NOx CO2 NOx SO2 Hg
factors listed for a variety of combustion systems. The (kg/therm) (g/therm) (g/therm) (μg/therm)
EPA provides only a single natural gas emission
5.281 4.14 0.026 11.4
factor each for CO2, SO2, and Hg.
64 Energy Efficiency (2010) 3:53–109

a
WA

0.063 ME
MT ND
0.106
0.074 0.061
OR MN VT NH
ID
0.065 0.066 0.1130.121
0.054 SD WI NY
WY MI MA RI
0.077 CT
0.062 0.144
0.078 0.124
0.062 0.117
0.115
IA PA NJ
NV NE
0.070 OH 0.085 0.106
0.095 0.060 MD DE
UT IL IN
0.079
CA CO WV 0.0900.076
0.061 0.078 0.066
VA
0.119 KS MO 0.055
0.076
KY
0.061
0.066 0.059
0.060 NC
TN
OK 0.069
AZ
NM 0.072 SC
AR
0.074 0.070
0.078 0.062 0.074
AL GA
MS
0.075 0.077
0.085
TX
LA
0.089
0.086

FL

0.082

AK

0.116

electricity price
$ / kWh
High : 0.190
HI
0.190
Low : 0.054

Fig. 6 Year 2005 average commercial sector prices in each US state of a electricity [de ðx; yÞ, $/kWh] (EIA 2006a) and b natural gas
[dg ðx; yÞ, $/therm] (EIA 2007a)

(EIA 2006b) describe a sample of 5,215 US commer-
cial buildings. The CBECS database was used to
characterize building inventory by US census division
(Fig. 3), since its records locate each building only by
census division.
The roof area of each building was estimated by
dividing its total floor area by the number of floors.
The database reports the exact number of floors in a
building only if it is less than 15. If the number of
floors was recorded as “15 to 25,” we assigned 20
floors (the range mean); if the number of floors was
recorded as “over 25,” we assigned 40 floors (a guess).
We calculated the adjusted conditioned roof area of
each building as the product of its roof area, fraction of
floor space that is cooled, and adjusted sampling weight.
The third term is the reciprocal of the probability of that
building being selected into the sample, adjusted for
non-response bias. The adjusted conditioned roof area is
the contribution made by the sample building to
conditioned roof area in the census division.
We estimated the CRA in each state by multiplying the
CRA in the census division containing the state by the ratio
of the state population to the census-division population.
National CRA is simply the sum of all state CRAs (or,
equivalently, the sum of all census-division CRAs).
CBECS assigns to each building one of 20
principle building activities (PBAs), such as “office,”
“education,” or “food service.” Using PBA and
age category (new or old) to define each building
class, we calculated for each combination of
census division d and building class j the ratio rj;d
of conditioned roof area to census-division popula-
tion (Table 5). A raster map of the CRA-to-LA ratio
uj ðx; yÞ for each building class j was created by
Energy Efficiency (2010) 3:53–109 65

b
WA

1.01 ME
MT ND
1.40
1.04 1.00
OR MN VT NH
ID
1.01 0.99 0.94 1.33
0.96 SD WI NY
WY MI MA RI
CT
1.00 1.01 1.25
0.91 1.39
0.89 1.29
1.26
IA PA NJ
NV NE
1.03 1.27 1.27
1.01 OH MD DE
UT 0.92 IN
IL
CA 1.13 WV 1.16 1.26
0.80 CO 1.08
1.09
VA
1.04 KS MO 1.19
0.91
KY
1.15
1.11 1.13
1.19 NC
TN
OK 1.25
AZ
NM 1.21 SC
AR
0.96 1.07
0.90 0.99 1.33
MS AL GA

1.17 1.27 1.43

TX
LA
1.02
1.11
FL

1.29

AK

0.48

natural-gas price
$ / therm
High : 2.47
HI
2.47
Low : 0.48

Fig. 6 (continued)

assigning each division’s per capita CRA for
building class j ; rj ðx; yÞ to the cells within that
division, then multiplying each per capita CRA
raster map rj ðx; yÞ by the county population density
raster map hðx; yÞ shown in Fig. 4.
Relating per-CRA energy savings and penalties
of building classes to those of building prototypes
The per-CRA cooling energy savings and heating
energy penalties of 10 office-like building classes—
new/old office, new/old laboratory, new/old non-
refrigerated warehouse, new/old public order and
safety, and new/old outpatient health care—were
assumed to equal those of the new/old office proto-
types. Similarly, the per-CRA energy savings and
penalties of 12 store-like building classes—new/old
food sales, new/old food service, new/old strip
shopping mall, new/old enclosed mall, new/old retail
other than mall, and new/old service—were assumed
to equal those of the new/old retail prototypes.
The per-CRA savings and penalties of new/old
religious worship buildings were assumed to be 40% of
those of the new/old office prototypes on the grounds that
the former are operated about 2 days a week, while the
latter are operated 5 days a week. Likewise, the per-CRA
savings and penalties of new/old public assembly build-
ings were assumed to be 60% of those of the new/old
office prototypes on the basis that the former are operated
about 3 days a week. Four difficult-to-characterize building
classes—new/old vacant and new/old other—were con-
servatively assigned zero per-CRA savings and penalties.
The ratios of per-CRA savings and penalties for
the 10 remaining building classes—new/old refriger-
66 Energy Efficiency (2010) 3:53–109

NEWE
14.5% c + 9.4% o +
NWPP 36.7% g + 27.6% n +
34.4% c + 0.3% o + 5.1% h + 5.7% r
10.6% g + 3.6% n + MROW MROE
49.0% h + 2.0% r 74.6% c + 0.6% o + 71.3% c + 2.4% o + RFCM NYUP
1.8% g + 16.0% n + 5.2% g + 13.2% n + 67.0% c + 0.9% o + 25.4% c + 6.6% o +
4.7% h + 2.1% r 3.9% h + 3.9% r 15.5% g + 14.3% n + 13.2% g + 27.1% n +
26.0% h + 1.4% r NYLI
0.0% h + 2.0% r 0.0% c + 58.2% o +
RFCE 35.5% g + 0.0% n +
44.9% c + 3.5% o + 0.0% h + 4.5% r
RFCW 9.6% g + 38.4% n +
72.8% c + 0.5% o + 1.6% h + 1.4% r
1.5% g + 23.2% n +
RMPA SRMW 0.7% h + 0.4% r
CAMX 80.6% c + 0.0% o + SPNO
13.5% g + 0.0% n + 84.7% c + 0.3% o +
12.6% c + 1.1% o + 78.1% c + 1.3% o + 2.0% g + 11.7% n +
46.4% g + 14.2% n + 5.3% h + 0.5% r
4.6% g + 15.2% n + 1.2% h + 0.1% r
15.1% h + 9.7% r 0.1% h + 0.5% r
SRTV SRVC
SPSO 65.8% c + 1.7% o +
2.4% g + 20.4% n + 51.0% c + 1.7% o +
58.8% c + 0.2% o + 3.8% g + 39.5% n +
AZNM 8.8% h + 0.9% r
34.1% g + 0.0% n + 1.7% h + 2.0% r
40.4% c + 0.0% o + 4.2% h + 2.3% r
31.5% g + 21.2% n +
4.5% h + 2.4% r
SRMV SRSO
23.4% c + 5.0% o + 64.0% c + 0.6% o +
39.3% g + 26.6% n + 10.1% g + 18.6% n +
1.6% h + 2.4% r 3.1% h + 3.5% r
ERCT
37.7% c + 0.5% o + FRCC
45.9% g + 13.2% n + 26.4% c + 18.3% o +
0.3% h + 1.0% r 36.5% g + 15.5% n +
0.0% h + 2.0% r

AKMS
0.0% c + 28.8% o +
3.6% g + 0.0% n +
66.9% h + 0.7% r
AKGD
12.3% c + 7.3% o +
68.0% g + 0.0% n +
12.4% h + 0.0% r
HIOA
18.0% c + 77.4% o +
0.0% g + 0.0% n +
0.0% h + 2.7% r
eGRID subregion electrical-generation resource mix
c = coal, o = oil, g = gas, n = nuclear, h = hydro, r = non-hydro renewable
HIMS minor "other" contributions omitted
3.6% c + 77.2% o +
4.1% g + 0.0% n +
3.0% h + 12.1% r

Fig. 7 Electrical generation resources mixes of the 26 subregions in the US Environmental Protection Agency Emissions &
Generation Resource Grid Integrated Database eGRID2006v2.1 (EPA 2007)

ated warehouse, new/old education, new/old inpatient Estimating state- and national-average saving
health care, new/old nursing, and new/old lodging— and penalty rates per unit conditioned roof area
to those of the new/old office prototypes were
obtained from simulations performed in an earlier US raster maps of per-LA cooling energy saving
study (Akbari et al. 1999). Specifically, the ratios of e^ðx; yÞ, heating energy penalty g^ ðx; yÞ, energy cost
savings and penalties for each building class to those saving c^ðx; yÞ, and emission reduction p^ðx; yÞ were
of an office building (new or old, as appropriate) were computed from Eqs. 17, 18, 19, and 20.
approximated by the mean ratios of savings and Per-LA rates of cooling energy saving, heating
penalties simulated in 11 US cities for each building energy penalty, energy cost saving, and emission
class to the savings and penalties simulated in the reduction were integrated over the land area bounded
same 11 cities for an office building. by each US state to compute state totals. Each state total
The matrices V and W relating per-CRA building (e.g., cooling energy saving in kilowatt-hours) was
class cooling energy savings and heating energy divided by the state’s CRA to estimate state-average per-
penalties to those simulated in the current study for CRA rates (e.g., cooling energy saving per CRA,
new office, old office, new retail, and old retail kilowatt-hours per square meter). US-average per-
prototypes are presented in Tables 6 and 7. CRA rates were computed by dividing national totals
Energy Efficiency (2010) 3:53–109 67

NEWE
NWPP 0.412
MROW MROE NYUP
0.418 RFCM
0.823 0.843 0.372
0.745 NYLI
RFCE 0.641

RFCW 0.497

RMPA 0.706
CAMX SPNO SRMW
0.923
0.399 0.894
0.837
SRTV SRVC
SPSO
0.678
AZN M 0.520
0.799 SRMV
0.569
0.515 SRSO

0.676

ERCT

0.644 FRCC

0.602

AKMS
0.218

AKGD HI OA
0.570
0.784 CO2 electricity emission factor
kg / kWh
High : 0.923
HI MS
0.661
Low : 0.218

Fig. 8 Year 2004 average electricity emission factor fe;i ðx; yÞ (mass of greenhouse gas or air pollutant emitted mass per unit energy
supplied to the grid) for each of four substances i: a carbon dioxide (CO2), kg/kWh; b nitrogen oxides (NOx), g/kWh; c sulfur dioxide
(SO2), g/kWh; and d mercury (Hg), μg/kWh (EPA 2007)

(sums of state totals) by national CRA (sum of state
CRAs).
State- and national-average per-CRA rates should
not be applied to specific buildings because energy
savings and penalties vary with both climate and
building type. However, these average rates can be
used to estimate and compare regional savings,
penalties, and reductions.
Results
Prototype cooling energy saving and heating energy
penalty per unit conditioned roof area
Annual cooling energy saving per CRA eðx; yÞ ranged
from 0.1 to 4.1 kWh/m2 for new office, 0.5 to
11.5 kWh/m2 for old office, 0 to 4.7 kWh/m2 for
new retail, and 0.8 to 15.0 kWh/m2 for old retail
(Fig. 9). New office cooling savings were at least
1 kWh/m2 everywhere except in very cold and sparsely
populated regions of Alaska. As predicted by Eq. 5,
old office, new retail, and old retail cooling energy
savings were about 3.4, 1.2, and 4.1 times those of the
new office prototype. Annual heating energy penalty
per CRA gðx; yÞ ranged from 0 to 0.104 therm/m2 for
new office, 0 to 0.235 therm/m2 for old office, 0 to
0.122 therm/m2 for new retail, and 0 to 0.264 therm/m2
for old retail (Fig. 10).
The dimensionless “load change” ratio ‘ðx; yÞ—
increase in annual heating load divided by decrease in
annual cooling load—calculated from Eq. 1 was less
than unity everywhere except in the aforementioned
cold and sparsely populated regions of Alaska
68
Fig. 8 (continued)
(Fig. 11). In most of the mainland USA, load change
ratios for new office, old office, new retail, and old
retail buildings were less than 0.5, 0.4, 0.4, and 0.3
respectively. That is, a cool roof almost always
reduced the annual cooling load more than it
increased the annual heating load.
Cooling energy savings generally increased south-
ward, while heating penalties and load change ratios
generally increased northward. Exceptions in the
mainland USA were induced by variations in local
annual mean global horizontal solar irradiance (great-
est in the southwest; see Fig. 1), local annual cooling
degree days (greatest in the southwest, Texas, and
Florida; see Fig. 2a), and/or local annual heating
degree days (high not only in the far north, but also in
cold mountainous areas; see Fig. 2b).
Energy Efficiency (2010) 3:53–109
Prototype energy cost saving per unit conditioned
roof area
Annual energy cost saving (monetary value of cooling
energy saving less monetary value of heating energy
penalty) per CRA cðx; yÞ ranged from −0.04 to $0.63/
m2 for new office, −0.04 to $1.72/m2 for old office,
−0.04 to $0.82/m2 for new retail, and −0.03 to $2.34/
m2 for old retail (Fig. 12). Energy cost saving was
greatest in Hawaii, which has high cooling-energy
saving, virtually no heating energy penalty, and the
most expensive electricity in the USA. The next
highest rate of energy cost saving (about half that
in Hawaii) was available in California’s Central
Valley. Savings were positive nearly everywhere—
even in Alaska, where expensive electricity and
Energy Efficiency (2010) 3:53–109 69

NEWE
NWPP 1.07
MROW MROE NYUP
0.57 RFCM
2.67 3.40 1.90
3.10 NYLI
RFCE 2.44

RFCW 3.64

RMPA 4.63
CAMX SPNO SRMW
0.92
0.25 2.75
3.15
SRTV SRVC
SPSO
3.28
AZN M 2.67
1.79 SRMV
0.65
1.04 SRSO

3.83

ERCT

1.44 FRCC

1.64

AKMS
0.31

AKGD HI OA
0.60
1.60 SO2 electricity emission factor
g / kWh
High : 4.63
HI MS
2.72
Low : 0.25

Fig. 8 (continued)

cheap natural gas made energy cost savings
comparable to those in central-latitude mountain
states, such as Utah and Colorado.
While energy cost saving generally increased
southward, there were large east–west variations
induced by wide differences in energy prices
(Fig. 6).
Present values of energy cost savings can be
obtained by using the PV multiplier b appropriate to
the roof’s lifetime and the real annual rate of return
(Table 1).
Prototype emission reductions per unit conditioned
roof area
Annual CO2 reduction per CRA pCO2 ðx; yÞ ranged
from −0.52 to 3.86 kg/m2 for new office, −1.0 to
11.5 kg/m2 for old office, −0.50 to 4.76 kg/m2 for
new retail, and −1.2 to 14.0 kg/m2 for old retail
(Fig. 13). Reductions were positive everywhere
except in very cold and sparsely populated
regions of Alaska with a low cooling energy
saving and a low CO2 electricity emission factor.
Outside of Alaska, reductions were lowest in the
northeast, which tends to have cool weather and/
or hydro power. CO2 reductions generally in-
creased southward, but were quite high in the
northern central states due to high CO2 electricity
emission factors.
Annual NOx reduction per CRA pNOx ðx; yÞ ranged
from −0.1 to 11.8 g/m2 for new office, 0.9 to 32.0 g/
m2 for old office, −0.4 to 15.3 g/m2 for new retail,
and 1.5 to 43.4 g/m2 for old retail (Fig. 14). NOx
reductions were positive nearly everywhere, with
70 Energy Efficiency (2010) 3:53–109

NEWE
NWPP 3.9
MROW MROE NYUP
4.4 RFCM
19.6 13.9 6.4
14.9 NYLI
RFCE 2.6

RFCW 18.7

RMPA 19.8
CAMX SPNO SRMW
7.4
1.0 12.4
18.5
SRTV SRVC
SPSO
11.5
AZN M 9.9
17.6 SRMV
11.5
5.0 SRSO

15.8

ERCT

13.2 FRCC

4.1

AKMS

AKGD HI OA
0.8
7.3 Hg electricity emission factor
µg / kWh
High : 19.8

HI MS
Low : 0.8

Fig. 8 (continued)

remarkably high values in the midwest driven by high
NOx electricity emission factors (Fig. 8b).
Annual SO2 reduction per CRA pSO2 ðx; yÞ ranged
from 0 to 17.1 g/m2 for new office, 0.2 to 45.3 g/m2
for old office, 0 to 17.4 g/m2 for new retail, and 0.3 to
61.3 g/m2 for old retail (Fig. 15). Reductions were
nonnegative everywhere, but much higher in the
east (except New York and New England) than in
the west. This strong variation was driven by SO2
electricity emission factors that are an order of
magnitude larger in the east than in the west
(Fig. 8c).
Annual Hg reduction per CRA pHg ðx; yÞ ranged
from −0.3 to 70.4 μg/m2 for new office, 1 to 193 μg/
m2 for old office, −0.2 to 75.6 μg/m2 for new retail,
and 2 to 262 μg/m2 for old retail (Fig. 16; reductions
were not computed in parts of Alaska and Hawaii
for which the eGRID database did not specify Hg
electricity emission factors). Reductions were pos-
itive everywhere except in sparsely populated
regions of Alaska with low cooling energy saving
and low Hg electricity emission factor. Variations
in Hg reduction were driven primarily by Hg
electricity emission factor, which is high when
coal dominates the generation resource mix.
State- and national-average rates per unit conditioned
roof area
Per-LA raster maps are not shown here because
spatial variations in the per-LA rates of annual
cooling energy saving, heating energy penalty, energy
Energy Efficiency (2010) 3:53–109 71

1.8
1.7 9
1.7 1.

1.9

1.4
1.4
1.4
WA
1.7 2.3 1.7 1.4

9
1.5 1.7 1.7

1.
1.
2.3

6
2.

1.5
2.2 1.7

1.5
1
1.8 1.6
2.3 1.5 1.6
2.0

7
2.2
1.8 ND

1.
2.0 MT 1.4 ME 1.7
1.4

2.0
1.5

1.7
1.6

2.1

1.7

1.5
1.9
1.8

1.5
3

1.8
1. 1.5 1.6 1.6

8
MN

1.
2.0 2 1.7 VT NH

1.8
.4 2.1 1.7 1.6 NY 1.9
3

ID

2.1
2.

2.6 1.9 2.
2.3

1.6

1.7
1.7 1.9 0
2.0
1.9

1.9
1.7

1.

2.0
1.7
1.7 1.8

1.9
OR 1.8

2.0
1.8

7
MA

1 .9
1.5 1.8 1.9

1.9
2.1
1.8

1.8
1.7 1.7 1.9 1.7 CTRI

7
SD MI .8

1.
8
WI

1.8
1.8 2.1 1

1.

1.6

1.6
1.7
2.4

1.5
WY 1.7

1.7
1.
1.7

1.
2.0

1.9
2.0
2.4

1.7
2.
71
1. .7 9

1.7
2.0
1.

1
1.9 2.2
2.7

2.8 IA NJ

2.0

2.0
2. NV

2.0
2. PA

2.0
3 1.8

2.1
3 2.2 2.3 NE
2

2.8 OH
2.

2.
2.2

3
2. 2.1DE
2.6

0
2.0

2.2
1 2. MD
2. UT 1.8 1.8

2.5
6 IN

2.1
IL 2.2
2 .6

3.2

1.9
CA 2.4 WV
3.2

7
3.1

2.5

1.
3. 2.4
3.0

2.7 1

2.2
VA

2.3
3.1 5 CO 4
2. KS

2.2
3.1 2.
3 2. 2. MO 2.3
3.

0 KY
2.

3.2

2.
2
3

4 3
2.
3.1

2.3 2.1 NC
3.1

2.
2.5
2.4

5
3.

2.5
AZ

2.0
2.7

0 2.8
2.5

TN

1
3.0

2.
2.9
3.0

3.2

2.8

2.7
OK 2.6

3.1
3. 3.1

3.0
6
7

3.
3. NM
3.

AR
.3

0
2 3.0
2.
3.5

5
3.3 3

3.
6

2.8 2.8 SC
3.4
3.4

2.7 AL 2.9 3
3.1 GA .0
3.2

7
2.
3.5

2.
2 MS 3.0
9 2.9

3. 3.3 3.0 3.1

3.1

3.0
5

0
3.1
3.2

3.
2.

3.1
3.6
3.3
3.0

.1

2.9
3.1 3.0
3.0 3

3.3
3.7
3.0

3.0
TX

3.1
3.1 3.8

3.2
LA

3.2
3

2.9
3.3

1
3.
8

.3 3.2
2.

3.2
2.8

8 3.1
2. FL
0.9
0.9
0.
7

2.7
0.7

0 1.0
1.
7
2.

0.6 2.6 3.
AK 2
1. 2.5
0.8
0.6

2
1.1

0.7
1.1

3.3
1.0

1.1
0.7
0.8

new office annual cooling-energy saving

0.9

1.0 2
2.9 kWh / m
0.9 .0
1 High : 4.1
3.3
HI
3.3
1.2

Low : 0.1

Fig. 9 Annual cooling energy saving per unit conditioned roof area ek ðx; yÞ (kWh/m2) for each of four building prototypes k : a new
office, b old office, c new retail, and d old retail

cost saving, and emission reduction are all dominated State- and national-average annual savings, penal-
by the four-order-of-magnitude range in US popula- ties, and reductions per unit conditioned roof area
tion density (Fig. 4), which swamps the one- to two- vary as follows:
order-of-magnitude range in per-CRA rates (Figs. 9,
10, 11, 12, 13, 14, 15, and 16). Instead, state- and & Cooling energy saving ranged from 3.30 kWh/m2
national-average values of per-CRA annual cooling in Alaska to 7.69 kWh/m2 in Arizona, averaging
energy saving (kWh/m2), heating energy penalty 5.02 kWh/m2 nationwide.
(therm/m2), energy cost saving ($/m2), CO2 reduction & Heating energy penalty ranged from 0.003 therm/
(kg/m2), NOx reduction (g/m2), SO2 reduction (g/m2), m2 in Hawaii to 0.141 therm/m2 in Wyoming,
and Hg reduction (μg/m2) are presented in Table 8. averaging 0.065 therm/m2 nationwide.
Also included for reference are state and national & Energy cost saving ranged from $0.126/m2 in
year 2004 census populations (millions), state and West Virginia to $1.14/m2 in Arizona, averaging
national roof areas (million m2), state and national $0.356/m2 nationwide.
conditioned roof areas (million m2), and year 2005 & CO2 reduction ranged from 1.07 kg/m2 in Alaska
state commercial sector prices of electricity ($/kWh) to 4.97 kg/m2 in Hawaii, averaging 3.02 kg/m2
and natural gas ($/therm). nationwide.
72 Energy Efficiency (2010) 3:53–109

6.

6.8
b
6

0
7.
6.4
4.4

5.0
6 .8
8.0 5.4
5.6

5.
4.8
6.2

8.2

2
5.0
WA

6.
7.8 5.6 5.4

0 7.
7.8

5.
5.

4.6 4.8
8

4
6
6. 5.8 8

5.

5.0
7.0
4.

0
OR MT ND 4

4.8
6.2
ME
0

5.8
2

7.

5.2
VT 6. NH
6.

5.6
MN

6.0
7.0 2
ID 5.2
6.4
9.2
9.0

SD

6.8

5.8
6.4

4.8

5.0
8.6 WI

6.6
7.8

6.8

5.4
6.6
5.6 MI MA

0
6.0
7.6

5.
7.2 7.4

6 .4
5.8 5.6

5 .6

5.6
NY 6.0 CTRI

5.4
6.0

6.2 6.0
7.0

5.6

5.

5.6
6.0
8.6 6.8

6.2
6
5.

5.4
5.
6.

6.
7.

4
6 0 6 6.4

6 .4

2
8.4

6.2 5. 0 5.
8
7.2 6.

6.2
9.8 6.2
IA NJ
WY 6.

6.4

6.4

5.8
8

7.2
PA

6.4
6.6
NE
9.0

7.4
8.4

NV OH 6.6

7.0
4 7.6 MD DE
8. UT 6.4 IL IN
10.
8.4
9 .4

4 8.0

6.0
4

5.
6.
8.2
CA

6.4
8
4

7.4
6.
9.8

6.8
7.6 8.2
WV

7.2
10

7.2

7.2
CO
8.
.0

10.2

8.8 KS MO VA

6.4
KY
.4
8.6

10

7.
8.
8.4

6
10.

9.0 6.8 NC
0

8.
7.8

8.2
8.
7.6

0
6.6 TN

8.8 6
9.2 8.6

9.0
7.
4
8.2
8.4

OK

6.
9.4

9.0

8.6
7.8

8.2
AZ 11 9.8 SC
.2
.0
9.2

10 NM
11

8.8
8.0

10 .2 9.6 AR
.8 9.2
10.4

10.0
0

9.0 MS
11.

8.8 AL 9.6 GA
10

7.8

9.
10.6

11.0

9.4
9.
.4

4
7.6

9.6
8.2

0
10.6

8.6

9.4
8.8

9.2

9.4
8.4 9. 9.2 9.0 8.8
8.6

9.4
TX 6
9.4
9.0 LA 9.2
8.4

8.6
8.4
8.6

9.
2 9.0
8.6
8.8
3.6

FL
3.4

8.4 8.8
2.

8.4
4

7.6

8.4
8
2.2

3. 4.0 8.
2
3.0

9.4 .2
10
0

8
8.

AK 4.2 7. 7.6

9.8
2.6

4.4
2.4

7.2
3.2

2.8 4.6
8.8
4.4
2.8

3.0
4.2
4.2
2.4

old office annual cooling-energy saving

4.0

2
kWh / m
High : 11.5
8.6
5.0 HI
8.8
9.0

Low : 0.5

Fig. 9 (continued)

& NOx reduction ranged from 1.70 g/m2 in New The present values of state and national energy
York to 11.7 g/m2 in Hawaii, averaging 4.81 g/m2 cost savings can be calculated using the PV multi-
nationwide. pliers in Table 1.
& SO2 reduction ranged from 1.79 g/m2 in Califor-
nia to 26.1 g/m2 in Alabama, averaging 12.4 g/m2
nationwide. Discussion
& Hg reduction (excluding that in parts of Alaska
and Hawaii for which eGRID Hg electricity Geographic differences in per-CRA rates
emission factors were unavailable) ranged from
1.08 μg/m2 in Alaska to 105 μg/m2 in Alabama, Spatial variations in the per-CRA annual rates of
averaging 61.2 μg/m2 nationwide. cooling energy saving and heating energy penalty
correspond reasonably well to those in annual cooling
and heating degree days. That is, the influence of cool
Each state’s aggregate savings, penalties, and roof installation on energy use maps well to climate.
reductions can be estimated by multiplying average However, per-CRA values of annual energy cost
per-CRA rates by the product of state CRA and the saving and emission reduction can depend as much
fraction of CRA to be made cool. National aggregate on regional differences in energy prices and electricity
values can be computed in a similar fashion. emission factors as on climate.
Energy Efficiency (2010) 3:53–109 73

2.4

2.5
1
2.5

2.
2.3

2.2
2.2
2.

1.9
WA

2.0

1.8
5

1.8
2.5
.5 2. 2.

2
2 2.0

2.2
8 1

2.
2.2

2.2
1.9

2 .9
2 .4

2
2.2 2.

2.

2.0
2.1
2. 7 1.9 2.1

2 .7
2. 7 2.

2.1

2.1
6 2. 2 2.2 2.2

2.2

2.0
2.3 7
MT

2.7
2.3
2.
ND ME
8
2.5

2.0
2.0

2.2
2.

1.9
2.
2.8
2.1

2.3
2.
1.8

2 .6

2.2
2.6
NH

2.1

2
2.5 3 MN

2.
2.3 VT

2.4 2.3
ID

2.5

4
.3 5

2.1
2. NY 2.4
2.9

2.1
3.0

2.4
SD 2.0
3.0

2 .2
3.2

2.1
2.

2.2
2.4 2.1 5 2.5

3
2.4
OR 2.

2.4
2.3

2.1

2.0

2.
2.1 WI 2.1

2.3

2.2
2.2 MI

2.3
2.3

2.1
2.0
2.3 MARI

2.
3.1 2.2 2.2

2.1
2.4

2
2.5

2.3

2.2
2.0
2.6 2.2 1.9

2.2
2.1
2.4
2.2 2.1 PA 2.1 CT
3.1

2.4
2.4

2 .3
2.

2.5
3.3 4 2.3

2.4
2.2 2.3
IA

2.3
2.3
2.5 2.7 NJ
3.5

2. 2.6

2.4
3. WY

2.6
3.0 NE 6 OH

2.5
0
3.7

2.9 2.9

2.
9
2.6 2.

2.4
3.1

2.8
2.4
5
2.7

NV .2 IL DE
2.6 2 3.3 3 2.5

2.4
CA 2.8 MD

2.4
UT

3.3
3. 2. 2.8
3.7

3 2.6
4.1 2.7
3.2

VA

2.3
2.6 IN
3.4

3.1 WV

2.8
4.0

3.1
2.7

2.9
CO 2.4
3.1

3.3
3.8

2.9
3.

2.8
3.4

3.0
3.3
9

3.1

3.0
KS

3.1
3.

3.4

3. MO

3.0
9

4.1 8 KY 3.
3.

4
0

3.
3.2 3.2 3. 3.2 3.5 3.3 NC
3.9

4
0

3.0
4.0

3.4 3.0 2.5

3.5

3.3
3.

2.5
3.0

2.8

3.4
3.2

1
2.4
TN

3.
8
2.

2.9
3.8 3.6 3.7 3.3 4.0

2.6
OK

3.7
3.3

3.4
AZ 4.6 3.9

3.9
3.5
3.7

4.5

NM 3

3.
4.0 3.0 AR 3.3

3.4
.6
3.8

7
4.4 4. SC

3.6
3.4
3 3.7
3.9
4.1

4. 3.4
4.4

4.5 5 4.0 3.5 MS AL 3.9

4.2

GA

4.0

3.6
4.0

3.5
1
4.2

3.4

3.8
4.0

3.7 3.8
4.
4.2

3.3
4.1

TX 3.8 3. 0
7 4.
3.6
3.7

3.8
3.5 3.5

3.8

3.9
3.7
.7 LA 3.8

3.8
3.8
3.7 3.6 3 7
3.
5 7 3.

3.9
3. 3.
4
3.

3.5
3.
3.7
3.6 3

3.9
7
3.4

3.5
FL
3.8
0.9

.7
1.
0.

1.3 3.5 3 3.8

5
1.1

3 .3 3.8
1. 3.5
0.7

3.7
3.3
AK 1.4 3.9
1 1.4
3.4

1.
4

0.6

4.2
3.
.9

4.2 3.3
0.80

4.1
1.6
0.4

3.2
5
1.

1.5
1.3
1.0
1.4

1.7 3.3
1.6

1.5
1.6
1.7

1.5
0.3

1.1
1.2

new retail annual cooling-energy saving

1.5 2
kWh / m
3.5
4
1.

1.5 High : 4.7

1.6 4.2
4.1

HI
4.3

Low : 0.0

Fig. 9 (continued)

While some of the regional variations in electricity conditioned roof area. For context, we note that the
emission factors can clearly be explained by genera- US EPA estimates that the typical passenger car is
tion resource mix—e.g., nuclear and hydro sources driven 12,500 miles (21,100 km) annually, emitting
produce no airborne emissions; coal and oil contain 5.19 t CO2 and 17.4 kg NOx each year (EPA 2000).
more sulfur than does natural gas; and coal is the The US EPA does not quantify vehicular emissions of
primary source of airborne Hg—emission factors are SO2 and Hg. Installing 1,720 m2 of cool roofing
not simply linear functions of fossil fuel mix. would offset the typical car’s annual emission of CO2,
Emission rates can be reduced through the use of air while 3,610 m2 of cool roofing would be required to
pollution control devices and efficient generation offset its annual emission of NOx. Retrofitting the
equipment. entire US stock of commercial buildings with cool
roofs would reduce annual emissions by 7.80 Mt CO2
Comparing emission reductions to emissions from (annual CO2 output of 1.50 million cars) and 12.4 kt
cars and power plants NOx (annual NOx output of 0.71 million cars).
Since cool roofs tend to save electricity at peak
The national-average annual rates of cool roof demand hours (e.g., late afternoon in summer), we
emission reductions are 3.02 kg CO2, 4.81 g NOx, also compare cool roof emission reductions to the
12.4 g SO2, and 61.2 μg Hg per square meter of emissions of a typical gas-fired plant used to produce
74 Energy Efficiency (2010) 3:53–109

9.
9.2
d
8
9.6

9.0 9.6
9.4 6.8

9.4

7.2

6.6
6.0
9.8

9.2

6.6
9.6

7.6

7.8
7.6 7.2 6.8

7.4
9.2 9.6

7.0
WA

7.4

7 .6
10 7.6

9.8
10.0 8.0

.4
7.8

7.2
.6

7.6
10

7.4
9.6

9 .2
ND 7. ME
MT 6

0
8.2 7.8
9.4

7.
9.2

8.8 4 .2
7.8
8.6 MN

8. 8
8. 6
10.4 7.2 7.4 0 8.
12. 8.8 NH

8.0
8 .0
NY

7.4 8.4 .0
VT

7.0
0 ID SD 7

7.6
7.0

4
7.4 .6 7.6
11.8

8.6

8.
8.2 WI 8
OR

7.6
7.4 6.8 8. 8.0

7.2
8.

7.8
8.2 0
7.

9.2
11.8 0

8.4

6
9.2 6 8. 8.4 7.
0 MA

6.6
7. 8.2

7 .0
7.4
7.4 7.8

7.4

7.2
9.4

8.0
RI

8.0
7.0
8.6 7.6 8.0 CT

8.0
WY 8

9.0
9 .2
7.2
7.0 7.2

8.
.8 10. 7.8 MI

8.2

2
0

7.8
8.6
8.0

7.6
9.8
12

10.2

7.8
13.0

7.0

8.2
8.6 8.0
.0

8.8 NJ

9.8
PA

9.0

7.8
9.6

7.6
NV

9.
9.8
8.4
12

NE 8.6 OH 8.2

6 9.4
IA 8.8 9.4
.6
.2

11.2
10

11.0 11.

10.0
12.0

12 8.4 MD
2 8.2
.6

.8
.0

UT 10. IN
.2
12.4
11

IL 9.8
11
10.0 DE

8.2
13

13.4 8
11

10.2
10.4
9.6

8 .2

8.6
.4
CA 11 9.4 WV VA

9.2

9.4
13.2
13.0

.0

9.2

11
10.0

4
10.
10.2

9.6

10.0
KS 8.4

10.
11.8

10
.6
11.
2

2 10
MO

.6
13.4
13.

0 KY

9.2
4
CO 11.
.4

12

10.8
11

NC
.6

12.0

.8
9.0
10.0

.8
12.8

11

12.6
.8

.8
12

11.2
12
11.2

9.2
.0
10

11
2 TN
11.
.2

.2

12

9.0
12

.0 12.4 9.2 OK 12.4 SC

11.6
AZ 12 12

11.8
4

.8 AR
13.8

NM .0
13.

11
13 12.6 .2

.8
12 .4
11

.2 GA
13.0

11
12.4 .6 12.0 12 12.8
.2

2
13. 13.6 AL 3.2

13
13

.6

14.6

12.
13.4 MS 1

.2
.0

12.

11.8
.4

10.4
10

2 12.2 . 0
.6
14

14.8

13

6
12 13.0
13.8

12
.6

12.4
.6
11.6

11

TX 13.
11.8

0 LA 12.8
11. 1 12.8 12.6
8 11.6 2.4

11.6
12 11.4

11
12.0 FL
11.6

.2
11.2
.0
12.2 12

12
2 .0 .8
4.6

.0
3. 12.4 1
2

0
5.

11.4

11.8
11.4
.4
5.6
3.2

10.0

11.2
12
5.0 4.8 4.2 2.8

4
5. 10 .8
.4 .8

2 .2 11
3.
4

5.8
4.

10 10

6.4
2.6

3.4

AK 9.6
5.8
3.6
6.0
4
6.

6.2
6.0

6.6

6.6
1.4

6.4
4.0

6.8 old retail annual cooling-energy saving

5.8

2
kWh / m
High : 15.0
8
11.
12.2

HI
Low : 0.8

Fig. 9 (continued)

peak power. In 2004, the average annual energy (annual CO2 output of 31.8 peak power plants),
output of the 1,380 US power plants that used natural 12.4 kt NOx (annual NOx output of 82.0 peak power
gas as their primary fuel was 471 GWh. For purposes plants), and 32.0 kt SO2 (annual SO2 output of 1,020
of comparison, we define a typical peak power plant peak power plants).
as a 350-MW gas-fired plant operated at a capacity
factor of 0.15 to generate 460 GWh/year, approxi- National sums
mately equal to the average annual energy output of
the aforementioned set of gas-fueled power plants. State and national potentials for commercial building
Using the average emission factors for this set of cool roof savings depend on both average per-CRA
plants, the typical 350-MW peak power plant would rates and the total CRA that can be made cool. For
annually emit 246 kt CO2, 151 t NOx, 31.4 t SO2, and example, retrofitting 80% of the 2.58 billion square
no Hg. Offsetting the peak power plant’s emissions of meters of commercial building CRA in the USA
CO2, NOx, or SO2 would require the installation of would yield
81.3 million, 31.4 million, or 2.5 million square
meters of cool roofing, respectively. Retrofitting the & an annual cooling energy saving of 10.4 TWh;
entire US stock of commercial buildings with cool & an annual heating energy penalty of 133 million
roofs would reduce annual emissions by 7.80 Mt CO2 therms;
Energy Efficiency (2010) 3:53–109 75

Table 5 Ratio of conditioned roof area to population (rj;d ; m2/thousand persons) for the stock of commercial building class j in each
US census division d

New Middle East North West North South East South West South Mountain Pacific
England Atlantic Central Central Atlantic Central Central

Vacant, new 0 97 33 24 54 20 217 64 61

Vacant, old 225 59 275 272 129 187 283 17 112
Office, new 433 538 834 643 674 540 859 654 597
Office, old 437 1081 755 1056 490 554 410 830 558
Laboratory, new 42 12 14 0 55 96 20 8 74
Laboratory, old 138 16 39 0 0 0 0 11 2
Non-refrigerated 138 567 428 517 673 532 797 403 273
warehouse, new
Non-refrigerated 331 181 764 690 365 135 365 233 166
warehouse, old
Food sales, new 124 75 178 171 159 61 313 33 50
Food sales, old 175 176 99 159 114 196 90 116 116
Public order and 9 122 50 219 110 118 170 81 18
safety, new
Public order and 88 63 23 36 9 38 19 37 55
safety, old
Outpatient health 24 57 159 166 207 170 126 264 98
care, new
Outpatient health 11 91 90 125 79 121 10 163 66
care, old
Refrigerated 0 0 1 7 39 34 14 0 45
warehouse, new
Refrigerated 0 3 68 0 2 0 0 0 12
warehouse, old
Religious worship, 0 158 273 296 459 616 298 245 80
new
Religious worship, 101 235 408 615 317 385 500 272 281
old
Public assembly, new 12 75 271 379 242 384 527 69 217
Public assembly, old 518 659 213 407 211 387 248 328 235
Education, new 36 71 554 353 1580 199 1262 852 349
Education, old 324 553 655 708 1049 844 998 976 981
Food service, new 23 53 155 129 279 122 375 73 184
Food service, old 122 66 172 389 182 286 194 40 92
Inpatient health care, 10 14 15 38 66 174 90 69 25
new
Inpatient health care, 10 68 81 105 137 46 36 49 57
old
Nursing, new 15 60 0 110 68 51 86 159 46
Nursing, old 50 24 119 382 126 50 8 132 51
Lodging, new 167 44 126 276 157 219 143 96 142
Lodging, old 204 114 102 459 102 324 99 311 175
Strip shopping mall, 169 126 175 68 393 216 444 178 207
new
Strip shopping mall, 267 88 200 2 146 141 195 195 87
old
Enclosed mall, new 338 63 0 48 187 46 52 166 179
Enclosed mall, old 20 68 42 109 0 28 123 78 5
Retail other than 178 181 231 490 566 908 409 294 486
mall, new
76 Energy Efficiency (2010) 3:53–109

Table 5 (continued)

New Middle East North West North South East South West South Mountain Pacific
England Atlantic Central Central Atlantic Central Central

Retail other than 333 350 399 432 379 464 473 370 392
mall, old
Service, new 63 202 126 407 366 350 171 192 66
Service, old 256 272 316 495 128 523 128 456 153
Other, new 19 71 55 6 172 0 65 4 37
Other, old 4 148 62 64 0 108 33 85 10

& an annual energy cost saving of $735 million;
& an annual CO2 reduction of 6.23 Mt, offsetting the
annual CO2 emissions of 1.20 million cars or 25.4
peak power plants;
& an annual NOx reduction of 9.93 kt, offsetting the
annual NOx emissions of 0.57 million cars or 65.7
peak power plants;
& an annual SO2 reduction of 25.6 kt, offsetting the
annual SO2 emissions of 815 peak power plants;
and
& an annual Hg reduction of 126 kg.
We compare these values to past estimates of
national savings (Konopacki et al. 1997) in Appendix,
finding good agreement after accounting for improve-
ments in building energy simulation and estimation of
building stock.
Summary
This study combines city-specific building energy
simulations, state energy prices, regional electricity
emission factors (mass of greenhouse gas or air
pollutant emitted per unit energy supplied to the
grid), and county-level estimates of building density
to characterize local, state-average, and US-average
values of cooling energy saving, heating energy
penalty, energy cost saving, and emission reduction
per unit conditioned roof area.
We used a building energy model with an
enhanced roof assembly heat transfer module to
simulate in each of 236 US cities the annual heating
and cooling energy uses of new and old office and
retail building prototypes. We estimated for each
combination of prototype and city the per-CRA
cooling energy saving (annual cooling energy use
with a conventional roof minus annual cooling-energy
use with a cool roof) and per-CRA heating energy
penalty (annual heating energy use with a cool roof
minus annual heating energy use with a conventional
roof) when roof solar reflectance is increased to 0.55
(weathered cool white roof) from 0.20 (weathered
conventional gray roof).
Annual energy cost saving (economic value of
cooling energy saving less economic value of heating-
energy penalty) was calculated using year 2005
average commercial sector prices of electricity and
natural gas in each state. Net annual reductions in
airborne emissions of CO2, NOx, SO2, and Hg
(emission decrease from cooling energy saving minus
emission increase from heating energy penalty) were
computed using year 2004 regional electricity gener-
ation emission factors (adjusted for transmission
losses) and non-regional gas furnace emission factors.
Weighted by conditioned roof area of the building
stock, state-average annual per-CRA cooling energy
saving ranged from 3.30 kWh/m2 in Alaska to
7.69 kWh/m2 in Arizona; the national average was
5.02 kWh/m2. State-average annual per-CRA heating
energy penalty ranged from 0.003 therm/m2 in Hawaii
to 0.141 therm/m2 in Wyoming (national average
0.065 therm/m2). A cool roof almost always reduced
the cooling load more than it increased the heating
load.
Annual per-CRA energy cost saving ranged from
$0.126/m2 in West Virginia to $1.14/m2 in Arizona;
the national average was $0.356/m2.
Annual per-CRA CO2 reduction ranged from
1.07 kg/m2 in Alaska to 4.97 kg/m2 in Hawaii
(national average 3.02 kg/m2). Annual per-CRA
NOx reduction ranged from 1.70 g/m2 in New York
to 11.7 g/m2 in Hawaii; the national average was
4.81 g/m2. Annual per-CRA SO2 reduction ranged
Energy Efficiency (2010) 3:53–109 77

Table 6 Matrix V of ratios of annual cooling-energy savings 105 μg/m2 in Alabama; the national average was
for each of 40 building classes to those of the four prototypes
61.2 μg/m2.
New Old New Old We calculated from a year 2003 survey of
office office retail retail commercial building stock the total commercial
building CRA in each of the nine US census
Vacant, new
Vacant, old
Office, new 1
Office, old 1 Table 7 Matrix W of ratios of annual heating-energy penalties
Laboratory, new 1 for each of 40 building classes to those of the four prototypes
Laboratory, old 1
Non-refrigerated warehouse, new 1 New Old New Old
Non-refrigerated warehouse, old 1 office office retail retail
Food sales, new 1
Vacant, new
Food sales, old 1
Vacant, old
Public order and safety, new 1
Office, new 1
Public order and safety, old 1
Office, old 1
Outpatient health care, new 1
Laboratory, new 1
Outpatient health care, old 1
Laboratory, old 1
Refrigerated warehouse, new 2
Non-refrigerated warehouse, new 1
Refrigerated warehouse, old 2
Non-refrigerated warehouse, old 1
Religious worship, new 0.4
Food sales, new 1
Religious worship, old 0.4
Food sales, old 1
Public assembly, new 0.6
Public order and safety, new 1
Public assembly, old 0.6
Public order and safety, old 1
Education, new 1.1
Outpatient health care, new 1
Education, old 0.6
Outpatient health care, old 1
Food service, new 1
Refrigerated warehouse, new
Food service, old 1
Refrigerated warehouse, old
Inpatient health care, new 5.5
Religious worship, new 0.4
Inpatient health care, old 3.1
Religious worship, old 0.4
Nursing, new 3.6
Public assembly, new 0.6
Nursing, old 2
Public assembly, old 0.6
Lodging, new 3.6
Education, new 3.7
Lodging, old 2
Education, old 1.5
Strip shopping mall, new 1
Food service, new 1
Strip shopping mall, old 1
Food service, old 1
Enclosed mall, new 1
Inpatient health care, new 16
Enclosed mall, old 1
Inpatient health care, old 7
Retail other than mall, new 1
Nursing, new 1.8
Retail other than mall, old 1
Nursing, old 0.9
Service, new 1
Lodging, new 1.8
Service, old 1
Lodging, old 0.9
Other, new
Strip shopping mall, new 1
Other, old
Strip shopping mall, old 1
Blank entries are zero Enclosed mall, new 1
Enclosed mall, old 1
Retail other than mall, new 1
Retail other than mall, old 1
from 1.79 g/m2 in California to 26.1 g/m2 in Alabama Service, new 1
(national average 12.4 g/m2). Annual per-CRA Hg Service, old 1
Other, new
reduction (excluding that in parts of Alaska and
Other, old
Hawaii for which Hg electricity emission factors were
unavailable) ranged from 1.08 μg/m2 in Alaska to Blank entries are zero
78 Energy Efficiency (2010) 3:53–109

a
6
04

0.0

0.054
0.

40
42

80
0.0

0.076
0.068
0.074 082
0 .0
0.05 0.0

72
07 0.070
4 74 0.086
WA

0.

0.0
0.0 0.07

74
74

0.0
0 8

0.0
5 0.076

4
0.074 0.0

0.0
.0 ME

84
0.0
0 0.070 86

78
0.074

0.
MT
0. 0.034

8
8

72
0.06
03

06
ND 0.08

0.080
0 .0
0.0

0.078
2

80
4 0.074
0.08
8

0.
30

2
88

66

0.0
0.08
0.0

0.0 064
MN
68

0.06 4

82
0.0 07 0. 72

0.
OR VT
70
0.0

58 0. 07 0

0.0
74
0.
0.052
6

6
0.0

0.
4
04

07
0.068
0.070

0.0
MI 07 NH
0.

0.
ID

0.0

0.0
6 0

74
74
0. SD WI 0.06 7 0 MA

0.0
0.058
07 66 NY 0.0

66
66
WY
0.0
8 0.0 0.062

52
0.080
0.068 0.068

4
0.070 CT RI

0.074
0

0.07

0.070
0.064 06

0.076
0.062 0.0

6
0.066

60 4 .
.06 0

62
0.0

60

06
0.0

82
0.07
56

0. IA 0

0.058
0.0

0.068 0.
74

0.0
0.050 07

0.0
NJ
0.0
0.0

0. PA

0.0
22 NV 0.0 0

0.052
0.0
07
0.020

64
0.0 2
64
6

0.0 0.058

72
40

4 OH 0.050

50
07

58
60 NE DE

0.0
0.

0.0
0

82

0.056
0 2 0
08

2 6 MD .050

0.058
0.0 0.0

62
0.0

80
64
0.

UT

86
68

0.0 IL IN

0.0
0.0
0.0

CA CO 58 WV 0.046
0.036 0.06
4 0.0

0.04
0.048
0.022 0. 0.0
80

09 KS MO VA 30
KY
2
0

6
0.0
0.09
0.02

0.048
0.044

4
0.046
0.014 0.05
6

02
2
0.058 0.04
0

60

0.
0.004 0.062 0.0 0. NC
0.020 0.018

0. 0
0.006

03 32
6

50 TN 0.034
0.010

0.0

0.0

0.020
54 6
OK 28
0.0
0.016

0.0
0.012

AZ 0.0 0
0.
4

30 AR

20
01

NM 8 SC
0.03 0.026
8 0.

0.0

0.014
0.020
16 0.032 0.022
12
0.01

AL
0.0

GA
4

MS
8
0.01

0.0
01

8
0.01
0.012

0.0
0.

14

TX 0.01 6 0.014
0.010
28

0.0 6 0.0 01
10 1 0.020 0.
0.022

0
LA 0.0 8
0.024

0.01
0.01
12

6
0.010

00
0.01

0.010

0.
0.002

2
0.018

0.
20 0.
0.0

00
01

8
0
78
0.080

FL
0.0

14 0
0.01

0.0
0.0
88

76 06

08
0.0 0.0 .004
0.0

74 0
AK 0.0

0.00
0.07
0 10
0.0

2
2 0.068
0.0

0.074 7
0.0
70

0.0
0.080

70

82
0.0
0.

0.0
new office annual heating-energy penalty
08

84
0

2
therm / m
High : 0.104

HI
Low : 0.000

Fig. 10 Annual heating energy penalty per unit conditioned roof area gk ðx; yÞ (therm/m2) for each of four building prototypes k : a
new office, b old office, c new retail, and d old retail

divisions. Local per-LA rates of saving, penalty, and energy saving of 10.4 TWh; an annual heating energy
reduction were estimated by multiplying a stock- penalty of 133 million therms; and an annual energy-
weighted average of per-CRA rates by the local cost saving of $735 million. It would also offer an
density of CRA (ratio of CRA to land area). We then annual CO2 reduction of 6.23 Mt, offsetting the
calculated state-average rates as the ratio of state sums annual CO2 emissions of 1.20 million cars or 25.4
(integral of per-LA rate over state) to state CRA. peak power plants; an annual NOx reduction of
National-average rates were computed in the same 9.93 kt, offsetting the annual NOx emissions of 0.57
fashion. million cars or 65.7 peak power plants; an annual SO2
We estimate that immediately retrofitting 80% of reduction of 25.6 kt, offsetting the annual SO2
the 2.58 billion square meters of commercial building emissions of 815 peak power plants; and an annual
CRA in the USA would yield an annual cooling Hg reduction of 126 kg.
Energy Efficiency (2010) 3:53–109 79

0.0 0.042
b

54
46
0.

6
0.0
15

46
0.08

0.1
40 6

0.116
0.1
0 0.1
0.1

96
5 58 50 0.1

4
.0

0.

48
0.1 80

15
2 0 0.1

15
.04

4
32

0.148
0.1
0.
44
0.1 60

14
1 0.196

0.1
0 .

6
0 0 0.

0.1 2
50 0.152

0 .1
16

0.

8
WA .11

02
MT 0.146

0 .1

18
0.1
0.056 062 4

20
8 2

58
0.15 ND

0 .1

0.
0.1
0.

74
ME

0.148
8
0 .1

0.134

84
6 0.1 0.1

46
86
0.156
0.162
8
0.1
0.102

26 0.13

0 .1
0.13 17 50

0.1 4
0.052

0.142
64
0.186

64

0.1
15
0. 70 0.162

98 2
04

6
68

80 0
6

0.1

0.1 18
0.1

0.1
0.1

0.
0.1
0.

0.13
0. 0.11

0.

0.18
22

15
VT

.
68

0.176
19
0
22
02 MN

56
0.0

4
0
0.12 0.126
0.09

.1
6
6
0.1

6
0.1 SD 0.1

0.1
4
48
0.0 17

14
4 NH

5
0.18

.16
ID WY

0 .1
18 56

66

30 0.13
82 0.

6
0.150

0.1
0 .1
0.148

0.144
OR NY

0.1
6 0. 0.102

60 0
72
0.134 .128 04
0.068

0 15

62

0.096
16 0.17 0.162

0.094
0.1

0.
0

2
2
8

12
0.

.13
15 0.14
6

4 2

0.14
0.14 MI MA
0.09

0.1

32

0.1
0.132

4
0.128 0.158

6 0
0.1

0 .1
0.130 2

0.162

0.150

0.1
4 72 12 CT
13

0.076
8 0. 0

10
0.0 0.126 WI 0.

0.1
0.

4
68

0.152

0.118 0.102
0.120

11
0

0.1
0.

0.1
0.10 8
96
0.098

64

6
1
26

0.08

0.1
0.0

34
0.1 22

11
0.154

0.0
16

.16 0.1
0 .1

6
4
IA

00
0.10
PA

0.

0.088
NJ
0.1

08
62 0.0

0
78

10
0.048 18

0.0
0.1
8
0.1

12
0.1 98 07

10
94

0.
0.1

0.
0.112 0.

12
4

0.
28 0

74
0.
0.164

0.0
0.2

46
06

0.
2
0.1 0.0
0.03

09

0.0
0.108
0.0 0.036

0.1

0.07
90 94 0.0

0.0
92
0 NE
0.

00

6
78

98
0.0
.1 2
17

98

0.09
0 .1
8

80
94
8

IN 92 MD
52

IL
4

NV
15

8
4

0.0
2

0. UT
0.1
0.15

CA OH
0.

12
56

8 WV

0.0
05 0.060 0.0 4
0.1

0. 0.0 88 0.1 0.124 8 0.082

0.072
0.05 11 0.050

70
0.1

86
VA

70
62 6 66 0.
0.

MO
4

0.0
0. 0.1

0.1

0.0
44
10

0.18
0.068

82
06 48
2

62
4

12

0.0
0.07

52
0.054 0

0.0
KS
0.1

0.044 CO KY 0
0

6 0.138

0.0
0.0
05

13 .07

60

82
2
0. 4
0.

0.024 04 0.07
4 06
0 .1
0.066 0.058

10 0.1 0. 0.044
0.1

4 0.1 0.0 NC
0.11
0.034

50

TN 72 0

0.
0.06
.04

38

0.046
08
OK SC
6

2
0.0
0.0

AZ 50 0 0.092
.07 0.064 AR 0.056 8
30

60

6 0.084 0.052 0.05 04

0.

0.
0. 4
0.0

0.048
6 054 NM 0.080
0.050

03
04 0.072 0.046 0.0
0.056

8 0.042

0
0. 05 0.040 30

0.0
AL
0.050

0. 0. GA
4

4
04 0.036 MS
05

0
50

8
0.

0.044 0.0
34 .032 20
0.0

38 0
0.030 0.034 6 0. 0 0.0
0.05

24
0.054

.03
44

0.020
0.0
0

6
0.0

LA

0.02
0.022
FL
2

14
0. 26 0

0.012
03

02

0.0
TX 02
0.022 0.0

0.0
48 0.
0.

6
0.0
20
0.0
20

18
0.2 0
0. 2

2
8

01
4
6

02

0.0
12 48
02

0.
15
0.1

0.1
0.

0.
4

46 0.140 0.
02

06
. 1 13
56 0 6 0.134
0.

0.0
8
02

0.002
0.178 0.1 3 0. 016
0.1 0.124 AK
0.
0.176
0.160

50 0.126

0.1
0.132

30
0.158

0.1
6
0.166 0.1 4
0.1

68
old office annual heating-energy penalty
0.158 2
therm / m
144
0. High : 0.235

HI
Low : 0.000

Fig. 10 (continued)
80 Energy Efficiency (2010) 3:53–109

0.018
8
00
0.

10

78
0.0
WA 0.0

0.070
68
0.034

0.0
82

0.070
0.0
0.090

0.034

84
0.0
02

0.0
0.006 0
08 0.1 ME 86

96
42

0.0
0. 0.0 0.0

0.0
MT MN

94
0.076 82 ND

88
0.090
32

0.076
0.034

0.0

0 .0
0.1 .096
0.002

92
78

70

0.086
88
0.0

0.

0.086
92
0.0 0.0
0.0 0.0

0.0
10
ID

06
0.0
36 94

06
0
0.06

0.
0.

0.098
0.0 0.092
8 98 0.0
02

09
0 .0
98
0.0
0 78 VT NH

0.068
0.082
0.

8
0.
0 .0

0 .1
0.0
0.066
WY

82
OR 0. 0
38 07 09
0.08
0.066

86
0.

02

0.0
4 SD 08 0 0.072 NY
4

0.062
MI MA

68
04

0. 0.070

0.0
WI 0.0

82
0.0 76 07
0.0

CT RI

0.0

70
0.072
6

80
80

0.0
0.084

0.06 0.0

54
0.0
84
0.038

0.07 0.06 0.0

0 0

0.0
4 8
04 .028

0.
0.0

IA 0. 0 58

0. 8
0. 0

2
0.0

06
48

0.
4 07 06

08

06
0.044

0.0 0 70 06 0.042 NJ

05
26

0.

2
NV .0 2
0.

0.

2
74

0.
.

62
30

0.06
06 0.068 PA

6
0.008

07
0 56

0.0

06
0.0
0. 68

0.04

0.074
DE

4
0.

0.0 0
6
0.022

0
0.0

0.
OH

0.
06

NE

04
42
0.05
07
UT IL IN MD
0.06

6
0.04 WV

4
CA 6 0.046

0.0
2

0.048
2 0.05 0.050 30
0.0

54
0.010 CO
0.0
KS 32
76

MO 0.0 VA

0.
60

03
KY
0.0

0.046
02

0.0

4
0.04 66
0.0

01
0.034 0.028
8 0.036 NC

0.
40
0.0
0

0.024
36
0.04
0.006
02

2 8 6 TN
0.01
0.0

01 01

0.004
0.
0.0

0.0

0. 0.
01

36
OK
14

6
0.002
0.004

AZ 0.0

0.
02

0.024

AR

00
0.0 NM SC

0.0
20
0.0

2
12
0.0

08
0.008 0.006 0.002
0.014 AL GA
0.002

02 MS
0.004

0.0

0.0
04
06

6
0.006
04

0.00
2
0.0
00
0.0
06

0.

02
0.0

TX
02

0.0
0.006

0.00

0.0

0.0
LA

08
8

6
.00
4

0 FL
00
0.084

0.
78
0.0

76
6

0.0
86
0.09

80
0.0
0.0

0.082
0.0 AK
82
0.076
2
0.0 8 0.07

0.06
0.

74
0.0
07

8
0
0.06

0. 88

7 2
0

0.0 new retail annual heating-energy penalty

09

8
0.06 2
therm / m
High : 0.122
00
0.0
HI
Low : 0.000

Fig. 10 (continued)
Energy Efficiency (2010) 3:53–109 81

d 80

0
0.0
06
0. 0.0
80 0.175
0
55 11 0.110 0.150
0.

0.100
0.0 0.

0.090

0.085
1 0.13
0.0 85
0 0. 30 0.1 35
10

5
0.12
0
0. 0.1

0.1
WA 15 05 0.125

0.120
55

20

0.0
0.0

0
25

15
0.115

0.1
0
0.0
0.070 0 ND 0.1 ME

95
11 90
MT 0.12
5 0.

0.1
0. 25

0.1

0.13 45
0.0
0. 0.1
11 35
60

0.1
0.1

05
11 0

1
NH

0.120
0.0

0.10
50 5
0.1

0
0.1 0
OR 0 0.125

5
0.1 .11

0.
2
50

45
0.1

0.1

14
0.10
0.135

0
10
ID 5 0.1

0.110
0.1
0.10

5
0.15

0
40 0.125 35 00 VT
SD .1

0
0.135 MN 0 WI

0.07
12
0
NY

0.0
WY 0 .155
0

MA

5
0.105
0.0

0.1
0.
0. 110
0.120

35

0.1 0
0.1

0.105

75
.19

0.
5
90
0.0

15 .105
25
55

0.1

10
40 RI

09
0.1
0

0.140 0.165

5
00

0.
05

0.0
90

0.055
90
12

0.09
0.1
0.1

0.115

0
0.120

0.125
0.2

0
0.0
0.15 CT
0.095

0 0.1 MI

0.0 85
0.145
0.055 0 0.1 25 0 0.
0. 0.1 1

0.140 0
11 5 0.11 0.075

0.1

0.0
80
09 07

0.135

0.0
.155

12

0.080
0.
0. 0 0.

0.080
35 0. 0

95
5
NJ

05

0.070

85
11

55
85
0.14

0.115
35 PA
0.07

95 0.0
0.060
5
65 0

0.

0.0
0.0

0.105

20
17 0.160 NE 65

0
0.085
0

0.0

.1

08
CA 0 OH DE
5

0.1
NV 0.175 IA
0.1

50

0.1
0.

0.06
75

55
0.155

0.0
UT 0.0 5

90
0.13

0.08
0.050

IN
80

00
0.0
0.

0
0.1 IL 40 VA

0.0
5

0.09
0.1 5
0.09
06

0.1
0.1

15
0.18

5
5
5 85 MD
0

50
0.155 0.185

0.1
0

0
0.0 0.0

0.06

5
0.0 180 WV

0.1
70
CO 0.

0.09
35

90
0.0 MO

00
0.0
0.1

0.0 KS

0.140
35 0.1
30 0.1 KY 0.080 0.05
0.

00

70

0
02

05
0.16 5 0.0 5
0

0.085

0.1

0.
70 5 0 80

0.0
0.010
0.1
3 NC
0.0

0. 0.135 0.105
16

60
TN
0.045 0.040
25

5
0

9
0.100
0.09

0.0
AZ 0 OK 0.
05 SC
0.0

0.020

0. AR 0
NM
3
05

07

5
0.055 0.025

0.070
0.0

0.06
0. 0 0.065
04 0.080 0.050 0.045
0 AL
25

0.0

5
0.0

MS

5
0

70 GA

0.030 0.045

0.0
0.03

02
0.03

0
0.04 0.

0.
0.0 075
0.045

40

45
20
0.0
0.0

40
0.035
LA

25
0.015
20
TX 0.025

0.0
0.0
0

0.0
25
0.020
03 0.0
0.01
0
0.

0.0
25

2
40

0
0.0 FL
95
05
0.130

0 5
0.00
0 0.0

5 02
0. 0.1

08 0. 05
0. 0.0
0.1 75

0.01 15
09

0.0
0.1
10

5
0.170 07 0.070
0

80 0.
0.0 5 AK 05
0.185

0.06 0.0
0
0.10
5

0.055
12

5
0.

.04
0
0

60
0.05
0.075
0.0
0.075

0.0
old retail annual heating-energy penalty
65

2
therm / m
High : 0.264

HI
0.
02
5

Low : 0.000

Fig. 10 (continued)
82 Energy Efficiency (2010) 3:53–109

a
6
0.3

0.
0.6

0.36
32
6

0.80
0.34 0.5
8
0.56 ME 0.84

0.60
WA

0.68
0.62

0
0.6 0.70

0.7
ND 4

0. 4 0.7
0.6
58

78
6 0.6

0.7
0
0.6 0.62
8

0.82
0.3

0.

8 0.
0.2

0.50
MT

0.5
0.50

0.6
0.74 NH

0
0.7
0.4

0.58
0.6
66
0.38 76

0
0.4

2
0.30 0.6 MN 0. 0.70

0
.56 0

0.52
0
0

8
0.26 0.58 0.52 VT 0
0.4 62
0.4

.54

0.6
0.50
0.42

0.36
0.
44

OR 0.48

4
0.50
0.

0.56

56
0.48

0.70
ID

0.62
0.5 6
0.5 NY
0.4
8

0.
4 0.44
0. MA
4 2 4

8
0.50 0.5 0.5
0.32

0.

46

0.4
SD MI
54

0.5
0.34 WI CT RI

0.56
0.
8

0.5
0.28 0.4 48
4
.

0.5
0.54 50

0.58
0 46

0.54
WY 4

4
0. 0.54

0.50
0.52 0.4 0.

2
0.52

52

0.4
2
0.3 0.3 6 0.4 PA 0.4

0.58
0.46

0.4
4

0.
4 6 . 0

2
0 2

6
0.3

0.5
0.5 0.4

0.4
0. 0 0.42 0.34
NJ
12

0.36
10 6
0.42 0

0
0.4 0.3

0.38
0 .4

NE
0.

2 0.
6 0.3
0.52

0.4 OH
NV IA 34
2
0.4

0.5 0
8
0.14 0
MD 0.30
0.4
0. 8

0.40
8

0.3

0.42
UT 50 IL
0.38

IN

2
36

0.28

0.5
CA WV
0.

0.36

0.26
0.28
CO 0.4 MO VA 0.18
KS
6

4 0.4
0.4
4

0.5 KY
0.5

6 0.30 6
0.1
8
0.30

0.3 0.34 6 NC
0.02 6 2 0.2 0.22 0.24

0.16
0.3
0

0.2
0.04

0.38
0.4

TN 0.18

4
32
OK

0.1
0.

0.16
AZ
0.06

0.08

4
NM SC
0.0

0.06
0.20 AR
8

0.1
4
0.0

0.12 0
0.14

0.08
0.06
4

0.08 MS AL GA
0.0

0.1

0.
0.0

1
0.08
0

0 6
6

TX 0.0
0.08
0.10

0.04
LA

2
0.04

0.0
0.10
2

1. 12 0.04
1. 1.2

86 1.
FL
1.78

0
14
1.3

2
8

4 1.0
1.7

28 1.0 0.96 0.02

0.06

1. 1.06
8 1.54

0.90
0.88 .94

0.

AK
1.70

92
0
8
1.26 1.24

0.9

4
0.8

0.8
1.3

6
0.7

1.
1.16

20
1.10
1.52

new office load-change ratio

1.08

[increase in annual heating load / decrease in annual cooling load]

1.24
06
1.

High : 2.0 (clipped)

HI
Low : 0.0

Fig. 11 Load change ratio (dimensionless ratio of increase in annual heating load to decrease in annual cooling load) ‘k ðx; yÞ for each
of four building prototypes k : a new office, b old office, c new retail, and d old retail
Energy Efficiency (2010) 3:53–109 83

0.18
0.1
0.12 0.5

0.34

8
2
0

0.7
0.4
14

0.
8

2
0
0.

0.50

58
0.4
52
0.1 0.38 ME

0.
WA 0. 0.50

0.62
0.60
0.3 54 0.7

0.64
2 0 0.46
ND

0.56
0.60

0.68
0
0.34 0.44

0.4
0.3
8

0.6
0.2
0.24 0.42

4
0.48

0.6
0.6
NH 0.48

6
MT

0.
58 0.52

34

0
0.

8
0.

0.40
2
0.36
OR MN

6
54
0.40 VT

0 .3
0.24

0.4
8

0.4
0.30

0. 0.4
0.2

0.5
0.

0.70
0.6
66 56 6 0.36

8
4
ID

8
0.46 0.34

0.6

2
0
0. 0
44 .44 0.42

0.3
0.6
WY 0.32 0.26

0.54
52 0.
0.40 WI 0.

0.4
0.3
2 MI
38 MA

0.44
CT RI

0.2 38
0.40

0.4
SD 0.40 56 NY

4
40

0.
0. 8

0.4

8
0.50 0.2

0.50
0.46

0.

0.

0.4

0.22
0.4 0.36
0. 0.36

46
6
8 0.22
.28

0.4
0.30

4
30 0.36

0.30
32 0

0.3
0.34 IA 0.18 NJ
0.28
0.10 0.44 28 0.

0.24
6

0.3
0.
0.32

0.30 0.

0
6

28
PA 0. 2 DE

0.
.48
0.3

8
0.18

4
0.3 NE

0.2

26

0.2
0.32
NV
0.08

0.

0.20
0.4 0 OH 0.2
8

4 46
MD
0.44

8
0.2

8
UT 0.3 IN

26
IL
2
6
4

0.
0.3

0.4 WV

0.34
CA 0.2
0
CO

0.16
0.2
0.1 MO

0.2
2 0.42 KS

0
2 KY
0.2 VA
0.08

0.12

0.

0.16 0.30

0.
38

12
0.06 0.28 NC
0.14
4
0.12
0.0

0.36
0.3
4

0.2
TN
0.0

2
8

0.10

0.16 OK
AZ
0.0

0.1 AR SC
8

NM
0.0

0.1 4
6

0.06
0
0.10
0.06
AL GA
0.08

MS
0.

0.08 0.1
10

0
06
0.0
4
0.06 0. 0.04
TX
LA
0.0

4
08

0.0
0.0
4
0.0
0.

2
FL
80

0.0

72
74 0.

0.
0.90

4
1.42

2
6 0.64
6

0.0
1.2

0.6
0.

82 0.5
0. 0 0.56 8
2

0.6 4
0.86 0.84 1.0

1.28 0.5 0.0

4 AK
4
0.5
0.46
6

0.52
0.62
1.0

0.88

0.7
6
8

1.00
0.6

0.
1.16

0.66

70
old office load-change ratio
8
0.4

0.68
[increase in annual heating load / decrease in annual cooling load]
60
0.
High : 2.0 (clipped)

HI
Low : 0.0

Fig. 11 (continued)
84 Energy Efficiency (2010) 3:53–109

Fig. 11 (continued)
Energy Efficiency (2010) 3:53–109 85

Fig. 11 (continued)
86 Energy Efficiency (2010) 3:53–109

Fig. 12 Annual net energy cost saving per unit conditioned roof area cðx; yÞ ($/m2) for each of four building prototypes: a new office,
b old office, c new retail, and d old retail
Energy Efficiency (2010) 3:53–109 87

Fig. 12 (continued)
88 Energy Efficiency (2010) 3:53–109

Fig. 12 (continued)
Energy Efficiency (2010) 3:53–109 89

Fig. 12 (continued)
90 Energy Efficiency (2010) 3:53–109

Fig. 13 Annual reduction in emission of carbon dioxide (CO2) per unit conditioned roof area pCO2 ;k ðx; yÞ (kg/m2) for each of four
building prototypes k : a new office, b old office, c new retail, and d old retail
Energy Efficiency (2010) 3:53–109 91

Fig. 13 (continued)
92 Energy Efficiency (2010) 3:53–109

Fig. 13 (continued)
Energy Efficiency (2010) 3:53–109 93

Fig. 13 (continued)
94 Energy Efficiency (2010) 3:53–109

Fig. 14 Annual reduction in emission of nitrogen oxides (NOx) per unit conditioned roof area pNOx ;k ðx; yÞ (g/m2) for each of four
building prototypes k : a new office, b old office, c new retail, and d old retail
Energy Efficiency (2010) 3:53–109 95

Fig. 14 (continued)
96 Energy Efficiency (2010) 3:53–109

Fig. 14 (continued)
Energy Efficiency (2010) 3:53–109 97

Fig. 14 (continued)
98 Energy Efficiency (2010) 3:53–109

Fig. 15 Annual reduction in emission of sulfur dioxide (SO2) per unit conditioned roof area pSO2 ;k ðx; yÞ (g/m2) for each of four
building prototypes k : a new office, b old office, c new retail, and d old retail
Energy Efficiency (2010) 3:53–109 99

b 5

6
4
4 WA
4 6

15
17

4
4 7 7

15
5 4 4 ME
ND

15
5

14
MT 13 18 14 13

15
17
14

4 11
5
26

14
22

22
9

7
11 1
20
26
13 VT
4

24
15
24

26

13
6 6 NH

0 1
MN 20

18
OR ID 13

9
16 16 19

6
7
14 NY

5 20

15
55
7 26

7 7
2 11
MI

6
5
6 SD 20

25

14
8

19

14
4
7 WI

19

98
MA

13
6 13
5 4 5 19 12

15
87

17

27 26
17
21
18 22 26 21 6

21 20
5
CTRI

7 6

19
6 7 22

4
WY

29
18 18

18
4 22 20

19

21
22

27
14

49
3 4

19
22

26
27
5

8
3

7 23 303 27 2
5

30
19 IA

28 34
6
57
31

17
6
5

32
15
3 PA 28

10
5
6

29
5 19
NJ

6
UT 31

20
NV

31
6 4

7
29

20
IN 33 32

30
67 7

16 8
6 8 29

9
8 22 22 33

19

21
7
19 20 19 34 34 OH
54

35 DE

29 29
28 27
8

38
7 68

25
20
8

79

22
5 NE 2 36

30
13
36 37 WV 82 38

29
19 23
3

IL 37

27
10 9 8 11 7 97
42 4 3

21
27 38
7 4

32

39
22 MD
9

20
CA CO 38

41
22
3

8
7

27

23
20
87 39 23
9

38
KS

39
7 MO

41
65

39
9

22
21 28 42 43 24 32 2

39
98

26

21
28
9

4 3

20
41 26 VA

24
9
3

KY
6

10

26

23

44
30 31
3

23
21 15 20 21
8

15 16 21

23
29 29 NC

25
24
9

9 14 23 17 9 10 24 25

15
3
6

12 31 27

25
25
12

25
30
7

14 13
6

6
7 12
6
1 17 TN 26
13

27
35 36

10
14
15

18
16

30
AZ 8
17

12

28
OK SC

36
7 13 AR 27 28
16

17

11
32

38 37 33
9

35
38 39
NM 16 14
7

13 14 13

11
17

14

22
30 19
40 4 39

29
37
15 AL
13

41
1
18 18 12
8

19
18

11
8

17 20

15

41
16

41
40
19
12 18 15
13

17 GA

36
8

40

19

11
12
40
17

7 14
12 16

TX 13 15

38
40 31
14

9
9 14 18 MS 3 38 20 38 37

13
11
13

40
13

38
15

15

18
1
13 12 0

17
34

18
12

16
16 LA

18
14 FL 16

15
14

16
17

11

17
14 14

16
1

13 17
18
2
AK 13 19
1 2

3
12

12
2
3

3
3

3
2

3
1

old office annual SO2 reduction

1

2
2
3

2
24

g/m
24
2 High : 45.3
27
HI
27

Low : 0.2

Fig. 15 (continued)
100 Energy Efficiency (2010) 3:53–109

c 1.4

1.4

2
1.
1.4 WA 1.8

1.4
2.2
1.2

1.4

3.8

5.8
1.4 2.4

6.2

6
4
6.6

5.
1.8

6.4
ME

6.0
1.

6.0
ND 2.6
MT 7.0

5.8
6.
3.0
9.4 9. 11

5.8
1.4

5.2
1.6

2.6
8
7.2
0
1.8

6.8 .2

9.8
1.2 10.2 2.8

8.8

3.4
4.4
1.6 2.0 2.0 MN NY VT NH
5.6 2.0 8

7.8
4.0 6.6 6.2
4.

3.0
ID

4.2 3.
8.2

2.4
4

6.6
6.

2.2

6.0
1.6
2.2

7.
OR

5.2
9.2

5.0
8.6
2.6

6
MA

4.0
8
WI

7.4
0.6

4.6
2.2

2
2.

9.2
4.4

2.2
MI 2.8 RI

4.6
0 .2
7.0
6.

4.2
0

7. 7

8
6.0

7.6

7.
6.8 .65.6

2.4
1 CT
WY .6 2.0 88 9.6

9.8

7.6
SD 9.

1.0
0.8

10
8. 9.2

5.0
NV 6.4 8.6 9.8 2 10 .0

7.0

8.0

9.8
.6
2.4
1.

2.2 2.6 11.6 .6 1.

8

11.4 11.6
IA 6.8
6

4
1.8

6.4
7.4

6.2
1 1 9

2.0
.8

6.
1.0 10.6 1 NJ

2.4
PA

.8
10
12.4
UT 2.0

10
0.8

4.2 4.4

7.4
7.6
12.4

3.4

.6
11
1.8

7.8
8.0

7.0 4.6
2.2

2.8 12.2 3.6

.4
.0

7.0
2.4 13.2 1 14.6

10
13 OH
0.8

7.4 8.0 8.4 8.4

9.0 9.6
2.6 2.4
1.2

2.6

8.2

.8
3.2
NE 7.2 IN 13.8 WV

2.6
2.4 14 8.6 MD DE

.4

8.8
0

IL

.4 12.6
9.2
8.8 .4 1
1.4

8.0
1.

13
2.8 6. 15.6 5.4
3.2 8

11.2
9.0
3.2
11.
2.6

CA CO 15.8 9.0 VA
1.6

15.0 4

10.0

10.4
2.6
3.

KS
2.

7.6

8.8 9.0 9.4

0

10

9.2
0
8

15.4
16. 16.6
1.8

3.2

3.4 5.6
MO 16.2

2
7.6
3.2 7.4
3.0

10.
11.0

7.
3.0 10.0 12.0 10

9.4
KY

.4
11.8

8.6
2.2

8.2
.0

8.0
3.

6.4 5.8 6.6 9.8

16
8
4

5.0 6.0 5.0 12

9.
2.2 3.2 2.2 9.6

.4
4.4 5.2 7.8 6.2 4.4 .8 NC

9.8
12
3.6

4.8
5.4 3.6

4
TN

10
.0
10.4

9.
11
4.6

11
12.2

.0
.0
5.2
SC 12.4

.4
10

10
12
6.2

OK AR . 12

6.26
AZ 2.8 8 8

14.2
5.

.6
.4
6 7. .8 14
3.2

3.8
NM .0
3. 4.8 4.8 12.4
0 5.0

15.0 14.4

.8
13.8 .6
3.8
6.4 6.

5.2

6.2
3.2

14
4.0
14

12
MS

4.6
6.6
5.0
5.8

6.0

.6
6.8 15.2 GA

14.6
.0
7.0

5.6
12

13
6

6.6 AL

16.8
15.4
6.0

2 LA

.0
6.2 7.2 15.8
5.

6.2
11.2
6.6

TX
6.0

15.2

4.8 9.2
5.8
5. 15.8

4.8
0
2 7.2 15.6 10.2

6
6.
5.0 4.2 4.8

15.
16 7.0

4.4
6.2
6.0 .0
5.6

5.2

9.4
4.2

5.4 6.8

6.6
5.4
6

5.2

4.4
5.

6.6
4.0

5.6
5.

6.0 6.8
8
0.4

FL
0.2

5.2
0.2

7.0
0.2 7.2
5.4

0.4

7.6
12.6
0.2

0.8
AK 5.2
1.0
0.6
0.8 5.8
0.8

1.0

0.
0.2
0.4

new retail annual SO2 reduction

0.6

6
2
10.4
g/m
High : 17.4
.4
12
HI
13.0

Low : 0.0

Fig. 15 (continued)
Energy Efficiency (2010) 3:53–109 101

d
6

7
6
6
6 8

13
WA

22
9
ME
5 21

21
22
6 MT 19 ND

23
31

14
15

24

40
38

34
VT

13
38
31
29
OR MN 32

14
10
76 17

28
ID 8 8 NY NH 9
18

21
8

27
24
24 WI

18
37
7

36
29

10
21 20 28 MA

26

17
16 15 8 20
SD

27

16
7 5 6 15

30
7 29 10 RI

24

16 3
23

28
6 MI
24 26 33

8
24 25 35 29 30 CT
WY

30
26
6 8 21 32

35

7
29 37

5
36

3
22

9
34 38 45

24
9
6

36 37

9 19
41 37 NJ
78
21 24

40

39
9

42
21
IA 41
9 6 28

40

35
41
8
NV UT 7 6

40
29

22

39

39
38
27

25
8 PA 39

19
10 42

41
10
8

45

26

30
10 27 47 OH 46 48

38 39
89

32 33 35
9 29 30 MD

28
4

50 WV 30
NE 24 25 26 49

12 25

39
11 IL 30
52

31
5

28

51
36
12 19 11 9

53
30
1 31 IN 53
9

52
CA

53
CO 1
10

26

31 36
33

31
54
56
9
9

30
29
11 11 52

28
6

54 56

59
38
KS VA

60
37

30
9

53
57

34
MO 31
10

32
12

28
7 55
KY 40 41 58 54 42 32
7

33

30
27 27 33 39
8 18 21 20 22 38 45 46
15 28 23 15

34

18
17 34

47

34
43
37

33
8
21 12 TN 36 35
9 18
9

16 16 NC

50
9
19
AR SC

44
AZ OK

45
10 18

13
35 51 46
NM

21
22

52 4
8

48

5
41

52 51 50
22 1
18 18 5 53
14

22

22
15
17 19

17
23 54

40 54
30
20

25 20 16
19 AL 55 57 56
GA
23

19

47
17
13 20 24 54

47
56
25

21

14 8
14

20
22

18

1 56
16 21

4
21 7

21
1 MS 54 42 23 34 37

36
17 18 5

18 1
54

47
16

49
23
14 21

47
TX

26
17
14

21
LA

23
24
15
13
23

22
15

22
20
1 FL

21
18 16
2
17

1 2 3
4
16

AK
4
2

4
3
33

3
old retail annual SO2 reduction
1

2
g/m
32
High : 61.3
36
HI
37

Low : 0.3

Fig. 15 (continued)
102 Energy Efficiency (2010) 3:53–109

8 8
a

5
WA

6
7
7 8

22
30

31
32

35
36
9 32
31 ME 6

31
10
OR MT 8 20 3 ND 30 29

30
9

31
10 23

13
6

27
33 29

37 38

26
28
7

9 11
VT

33
26

33
8 9 35 MN 34 12
WI

8
25
NH

8 8

28
11 12 ID 11
26 13 37 24
37 25

15
13

34
12 14 15

5 16
40 NY

13

27
SD

33
14

7
29
11 MA

41

10
43

33
12

8 13 1
37 MI

13
8 9

12
11 36

2
35
39
7 RI

28
26
36

34 3
9
39 734 1
33 16 2 26 6

26
2

38
WY 15 32 CT

29
31 35

30
40
8

36 38 5

65
40 3
36 PA 42

30
33

9
16
10

IA 30 38

36
46
41 40 41

10 14
11
13

36
15

40

39
36 NJ

14

38
42
3 42 44

37
NV

43
12 39

41
32

43
16 35
37

42
40

39
16

18 41

41
17 NE 38 OH 46
27 41 42 43 DE
14 16 15
34

47
10 39
12

26
2

2 19 4 42
44

24 26 23 22 15
UT 37 31 35 28 33 32 34
47 462

16
IN

44
41 IL 41 6 MD
20 45 4 48

41

47
19

31
5

22
CA 48 47

33

51
17
10

24
CO

49
19 24
WV 52
25

51

27
49
67

50

29
39 18 51
32 KS MO 28 VA 25

40
46
20 55 53
KY 29 1 27 50 26
8

47
32
38

34
30

44
3

56
40 36 37 46 27 26
11

30 29

47
28 31 47 19 42 21 NC
34
30

31
13

28
43 28

39
32
38

22 32

28
29

38
35 49

17
33

TN
38

37 47 33 30
43 43
14 40 39

39

51
39

29
52
41

54
37

33

32
48 41 45
50

AZ
38

OK 45

42

46
44

42

33
48
50
49

NM 36

16
34
46

AR
38

40

40
35

30
44 5 47
42

50 46 46 50 30 SC

51 48 47
45 43

44

38
44
52 55 57

15

21

28
3 7 GA

39 18
49

41
40

35
53 54 42 3 46 0
38 AL 5

52
52
56

54

55 30
39
38 53 49 41
51

45

54
55
47 53
38

58 59

56
48 MS

18
49
47

44
45

23 60
20
46

52

4 20
38
46

2 26 34 26 57

17

65
18 5
44

40 36
54 53 52

44
47

56
16

18 5
47 15

16
TX 40 38 23 17 61
LA 23
40

15
41

18
39
14

25
FL

39
38 7
3
AK 36

0
0
0 0

new office annual Hg reduction

2
µg / m
High : 70.4

HI
Low : -0.3

Fig. 16 Annual reduction in emission of mercury (Hg) per unit conditioned roof area pHg;k ðx; yÞ (μg/m2) for each of four building
prototypes k : a new office, b old office, c new retail, and d old retail
Energy Efficiency (2010) 3:53–109 103

b 32

32

32
32

24

18
34

34
30 26

38

42
WA 20

106
26

11
30 120 8 22
10 ME

32
26

106
8
32 ND
12

24
MT 96 102
6

32

118

100
OR

128
28

96
100 10

10
102
78

13
4

40 38

24
86
124

114

40
110
30

98

0
10 100 VT

90
10
6

36
108 76
36 WI 42 NH
36

8
28

0
116 112 MN

34 30
44 96 46 40 116 90

40 46
ID

11
44 42
40

4
7 NY 26

86
11 6

10
46
34
42

78

22
110

44
44
SD 0 28 MA

90
92
2
94

32
42

44

88

40
44 5
32 36 30

6
42
MI

50 48
30

10

106
30
8044 104

0
90 98 CTRI

94

28
12

88
34
12
4 48 56

28
32
13 118
130 4

108
1105

114
42 6

120
16

96
90 11

86

102
38
30

14 120 84

58
6

54
50 120

104
38

106

126
2

4
52
IA 12 8 136

0
134

13

14
11
54

34
38 48
126 12 2 1
40

12
14 120 13 1 28
2 NJ

6
WY 14

8
13
138 PA 14

2
NV 0

8
134
42 8

44

8
6 3

13
4
12

6
2

74

14
0
10

138 1

13
44 58 60

90
138
140 150 154 1
58

8 70
134
34

46 52 56 54 NE 110 128 OH 34

146
DE
30

146 152 WV 80
42

62
UT 68 112 90 86 94 98 92

82
IN 80 8

14
32

132 IL 160

162
0
72

16 2

110

6
8 8

76 88 78 74
158 160 92

78
66
CA
28

94
MD

176
92
86

170
52
4 6

82
16 17 VA

94
38

86

4
10

172
122 CO

14
64

168
KS 162 96 96 80
82

66 MO 14 18
174 88 8
6
8 98 KY 6 166

88
126

6 84
84

154

140
4 10
10

100 120 102 96 154 10 0 10 NC

92
10
80

158
114
118

4 10 90
46 114 70

144
6 8 4
124 2

88
108 0 10 TN 0 0

130
15 66 1 1
106

90
OK 126
122

11

116

92
44

52
0 146
2

98
122

11
12 14
0 104 4 96

8
12

16
8
2

16
26

94
10
158

13

6
14 SC

6
AZ 11
132 110

15
114
13
122

1 1

4
8
2 3 4 98
152

8 36 14

164
15
13
140

NM

2
154

0 122 146 AR 11 0 156 4

13 150 2 48
22 14 160
164

158 154
8 100 16

2
114 1

74
2

11
15
134

64
14 0 6
118

90
12
172

6
16

16
8 18 AL

170
120

11 13 17

60
4 54 6

168
14

6
136

124

140 150 2
116
132

48 MS GA
0

124

50
134

180

64
70
52 16
146

TX 12 182
2 6 160

68
2

130
8 16 66 114
15 156
12
10 78 66 170 62

76
128

122
4
2 1

180
6 54

98
12 2
8 60 54
42

176

54
74 48 174
130

72
132

8
12 LA 7 4
126

42 0

94
8 52

56
130
124 122 FL

40
120 112
8

6
11

4 44 4
11 8
AK 10

2
2

old office annual Hg reduction

2

2
µg / m
High : 193

HI
Low : 1

Fig. 16 (continued)
104 Energy Efficiency (2010) 3:53–109

c
10

11
11

11
8

9
12

7
WA 46
9
10

20
ND

11

45
41 8

44
10 48 42
ME

47
12
11 39 50 42

42
49 32

46 45
MT

41
13
47

48
51

48
49
51

40
10

14
43 32

12

35
44
45
12 48

9
WI 33

36 35
16 ID 28 15 MN VT

49

31
13 15

43
NH

12 18
15

38

19
14

11 15 10 9
17

31
11

43
47

46

37

8
44
OR

16
46
14

34 9
11

13
45
NY 13

36
11

15 18
MI

16

16

10
14 14

38
MA

42
10

45
30 8

36
38 31 23 28 CTRI

44
18 41 4
3 WY 11 SD 43 29

35
38

50

37
39
42 36

43
11 50

21
19

51
16 6

37

7
49 4 47 53
11

52 NE IA 49 42 44

52
15

48
15
50

17

20

46
14 21

37 39
17 50 47

52

45
54 NJ
NV 16 25 56 50 PA MD

56
18

42

56
23 OH
17

17 47

51
51

53
4 53 52 61 6
0 60

54
19 53 52

52
DE 54

61
42 50 IN
19 18
5 58
3

26 2 26
16

3 49 56

25
32 5 2
61

UT 47 40 46 41 44 43 45
IL 0 63 6

25

52
21

57
31
59

62
47 51
37

66 65

36
28

64
4

35

30 32 31 29
30 33 32

22
1

1
47
6 6

64
14

CA CO

22

68
35 6
65 3

24 21

32 32
58
55
34 35

72 VA
24

68 WV

62

71
20

67
37
68 36

66
KS
8

41 69

50 36
25 MO 38 70

34
11

54 53 57

32
9

41

40

38
39 KY

70
42 49 43

37
33

37 39 38 61 NC
41

40 40

60

35
33
4 27 35
44

57 21 52

44
OK 46 42

39
38

37 38

36
47 17

48
25
37

45
53 TN 36
56
AR
47

58 4 41 39

51
45 18 52 50 43
53 5

61 6

59
48

41
10 47

64
42 43

5
55 54
61

AZ

45
57 55

46
48
50 26 43 49 44 SC

18
40
56

19
NM 59
58

19
57
58 60 55

7
MS 39
57

51 44

40
59
57 4
46 23

62 60 56
55 58 47 62 2 7 57
62 4
59
54

49
45 61
52

21
23
46 53

56

44
63
6

47

AL 63
61
69

GA

37
53
56

64 26

55
51
57

66
62

19

67
58

70
60
67

48

55 54 60 6

69

24
49 65 6 69

67
59

TX 6853 59
55

64

65 60
63 26
48

58 26 61
49
48

61 71 65 66
52 46

6 35 23
50 52 2 72

35
6

17
70
39 75 6
17

57
73
46

27

4
30
51

58 6
54

74
19

53
53 5

56

21
55 LA FL
4
56

51 54
48

17
50 18
49

19
47
48

AK

26

new retail annual Hg reduction

2
µg / m
High : 75.6

HI
Low : -0.2

Fig. 16 (continued)
Energy Efficiency (2010) 3:53–109 105

d
45
40

45

25
45

35
40
45
WA

95

165
35
30

160
35

15
ME

160

155
50

0
35
45 MT 140
115

17
115

105
50
140

90

0
175
135

45
160
OR ND MN

0 11 20

35
115 55 50 165 VT NH

1
ID

65
NY

130
45 60

120
SD

115

40
60 30

160
50

120

135
130 150
70

75
60
55

WI

125
MA

180

55
40 135 55

170 1
40
40

35
65 60

12
45 15 15 50

125
40
5 0 14 65 35

50
175 MI 0 5 100 90 RI
15 5 60 12

35

13
0
20 CT

0
WY 65 155 0 16 15

12

19
180
185

5
155

80
40 0

40
45 60 165
40

70 NE 5 150 17 0
19 15 19

210
IA 5

45

55 65
185

85
65
0 175 NJ
165

20
65

175
15 45
45
NV 17 OH PA

5
180
110

180
195
45
0

13
80 185 17 0 200

80
75
5
0 IN 19

0
0
50

60

17

195
7

75
135 120

215
60 75 65
170 1 5 5
35

0
220
0 19 21

90
MD

225 2 1
90 140 160 145 155 20

75
UT

90
220 2
40

0
IL

115
160 110 DE

18
0 30 115

125
100 85 110105
5135

75 5 15
30

CA 11

175
15

CO 220 WV VA
45

125

23
11

250
130 11

255
240

115
130 11 0

225

5
9

KS MO 245
0

12 5

205
15

225
5

0
95

KY
14

140

5 250
5

120
13 120

12
15 165 125
70

0
170 175

205
20
165

5
145 135 85 NC 125

160
195 65185

185
5

0
140
13

12
75

15
70
0 5 205 TN 130
15 0 145

5
15 205
18 60 5

17
25 155 19 SC 135

5
OK
150
AZ

18
175

18
155
0 125
20
14

AR 15

215

5
NM

65
210 0 135

210
0 2 225

215
195

205 155
160 210 215 190 85
20

140

10

155
0

16
15

17 5 225 230
230 2

220

0
16
170

160

0
240

110 AL

22
80
190 0

230
155

85
12

235
190 0

0
18 215

175
200 16 225

125
80
0 235 190 MS GA

180
TX
5

10
160
20

00 70 200 120
170

0
10 100 115 2
170

LA

100
140
5
180

90
70

95
85 22

65
85
85 0

65

155
175

165
180 6
5
5 FL 5
180

17
18
5

165 50
1
155

0
15 0
AK 14
5

old retail annual Hg reduction

5

2
µg / m
High : 262

HI
Low : 2

Fig. 16 (continued)
106 Energy Efficiency (2010) 3:53–109

Table 8 National and state average rates of commercial-building annual cooling energy saving, heating energy penalty, energy cost
saving, and emission reduction per unit conditioned roof area

Region Population Roof area Conditioned Electricity Gas Cooling- Heating- Energy- CO2 NOx SO2 Hg
(millions) (million m2) roof area price price energy energy cost reduction reduction reduction reduction
[CRA] ($/kWh) ($/therm) saving penalty saving (kg/m2 (g/m2 (g/m2 (μg/m2
(million m2) (kWh/m2 (therm/m2 ($/m2 CRA) CRA) CRA) CRA)
CRA) CRA) CRA)

USA 296 3930 2580 – – 5.02 0.0645 0.356 3.02 4.81 12.4 61.2
AK 0.666 6.66 4.56 0.116 0.48 3.30 0.128 0.319 1.07 5.94 1.91 1.08
AL 4.53 59.0 43.8 0.075 1.27 6.33 0.0372 0.421 4.56 7.04 26.1 105
AR 2.78 45.7 29.6 0.062 0.99 4.69 0.0526 0.234 2.77 3.91 6.44 40.7
AZ 5.75 73.9 49.4 0.074 0.96 7.69 0.0517 0.512 4.62 7.90 5.62 96.9
CA 35.9 359 245 0.119 1.04 6.13 0.0292 0.699 2.58 2.31 1.79 8.18
CO 4.76 61.2 40.9 0.076 0.91 5.48 0.133 0.296 4.92 8.08 5.61 43.5
CT 3.48 35.6 18.8 0.115 1.26 4.96 0.0809 0.468 1.84 2.11 5.89 20.3
DC 0.567 8.28 5.94 0.091 1.28 3.81 0.0824 0.241 1.68 2.98 14.7 72.3
DE 0.831 12.1 8.70 0.076 1.26 3.84 0.0799 0.191 1.70 2.97 15.6 79.0
FL 17.5 255 183 0.082 1.29 5.72 0.0115 0.448 3.77 6.45 11.1 29.7
GA 9.05 132 94.8 0.077 1.43 5.39 0.0482 0.341 3.79 5.72 22.9 93.7
HI 1.25 12.5 8.54 0.190 2.47 6.02 0.00304 1.14 4.97 11.7 12.8 34.2
IA 2.98 50.7 32.4 0.070 1.03 4.66 0.105 0.214 3.71 8.52 13.8 101
ID 1.39 17.9 12.0 0.054 0.96 5.20 0.118 0.168 1.79 3.77 3.31 24.1
IL 12.8 179 110 0.078 1.09 4.22 0.0994 0.217 2.97 5.48 19.6 89.9
IN 6.29 88.1 53.8 0.066 1.08 4.72 0.0849 0.215 3.25 6.41 24.3 103
KS 2.77 47.1 30.1 0.066 1.11 5.23 0.0855 0.250 4.74 10.1 15.9 71.1
KY 4.16 54.2 40.3 0.060 1.19 5.20 0.0709 0.228 3.57 6.61 20.3 74.0
LA 4.59 75.5 48.9 0.086 1.11 4.83 0.0197 0.389 3.08 4.20 6.68 45.1
MA 6.45 66.1 34.9 0.124 1.39 4.68 0.0874 0.460 1.68 1.94 5.56 19.0
MD 5.54 81.0 58.0 0.090 1.16 4.20 0.0891 0.270 1.99 3.60 17.6 86.5
ME 1.32 13.6 7.17 0.106 1.40 4.56 0.115 0.323 1.48 1.77 5.42 18.2
MI 10.2 143 87.3 0.078 0.91 4.13 0.101 0.230 2.88 4.74 14.5 68.2
MN 5.22 88.6 56.6 0.066 0.99 4.17 0.137 0.136 3.09 7.45 12.4 89.5
MO 5.78 98.2 62.7 0.059 1.13 5.50 0.0832 0.230 4.77 7.95 18.2 99.3
MS 2.92 38.0 28.2 0.085 1.17 6.28 0.0359 0.485 4.10 6.21 18.4 73.7
MT 0.918 11.8 7.90 0.074 1.04 4.74 0.135 0.211 1.58 3.56 3.50 25.2
NC 8.70 127 91.1 0.069 1.25 4.91 0.0604 0.258 2.52 4.37 14.6 53.5
ND 0.644 10.9 6.99 0.061 1.00 4.13 0.126 0.126 3.11 7.42 12.2 88.7
NE 1.77 30.0 19.2 0.060 0.92 4.79 0.0939 0.197 3.90 8.77 13.9 101
NH 1.32 13.5 7.13 0.121 1.33 5.35 0.121 0.482 1.82 2.14 6.36 21.6
NJ 8.71 101 60.1 0.106 1.27 4.72 0.0786 0.400 2.14 3.61 18.5 92.7
NM 1.92 24.6 16.5 0.078 0.90 6.92 0.0921 0.456 4.15 7.14 6.13 91.0
NV 2.37 30.5 20.4 0.095 1.01 6.86 0.0737 0.570 3.64 6.37 4.74 71.8
NY 19.3 224 133 0.144 1.25 3.80 0.0732 0.452 1.37 1.70 5.59 17.5
OH 11.5 161 98.2 0.079 1.13 4.45 0.0808 0.260 3.05 6.02 22.8 96.8
OK 3.57 58.7 38.0 0.070 1.07 3.83 0.0646 0.195 3.06 4.66 7.62 73.9
OR 3.58 35.8 24.5 0.065 1.01 4.83 0.0589 0.254 1.93 3.71 3.07 22.9
PA 12.4 144 85.8 0.085 1.27 4.69 0.0830 0.289 2.40 4.31 20.2 98.0
RI 1.07 11.0 5.80 0.117 1.29 4.61 0.0730 0.445 1.71 1.95 5.43 18.8
SC 4.23 61.7 44.3 0.074 1.33 5.38 0.0466 0.330 2.86 4.86 16.0 58.9
SD 0.785 13.3 8.51 0.062 1.00 4.47 0.122 0.155 3.49 7.92 12.3 89.6
TN 5.93 77.1 57.3 0.072 1.21 5.78 0.0588 0.339 4.04 7.31 21.1 73.5
TX 22.9 376 243 0.089 1.02 4.93 0.0253 0.408 3.39 2.72 7.72 70.7
UT 2.41 31.1 20.8 0.061 0.80 5.43 0.116 0.233 2.03 4.13 3.54 25.9
VA 7.51 110 78.6 0.061 1.15 4.35 0.0789 0.170 2.26 4.15 14.6 55.8
Energy Efficiency (2010) 3:53–109 107

Table 8 (continued)

Region Population Roof area Conditioned Electricity Gas Cooling- Heating- Energy- CO2 NOx SO2 Hg
(millions) (million m2) roof area price price energy energy cost reduction reduction reduction reduction
[CRA] ($/kWh) ($/therm) saving penalty saving (kg/m2 (g/m2 (g/m2 (μg/m2
(million m2) (kWh/m2 (therm/m2 ($/m2 CRA) CRA) CRA) CRA)
CRA) CRA) CRA)

VT 0.630 6.46 3.41 0.113 0.94 4.95 0.112 0.454 1.67 1.97 5.90 20.0
WA 6.19 61.8 42.3 0.063 1.01 4.46 0.0602 0.220 1.75 3.40 2.84 21.1
WI 5.55 77.8 47.5 0.077 1.01 3.76 0.128 0.157 2.58 5.46 15.9 71.2
WV 1.82 26.6 19.1 0.055 1.19 4.33 0.0941 0.126 2.90 5.80 22.2 94.2
WY 0.502 6.46 4.32 0.062 0.89 4.86 0.141 0.171 2.65 5.00 4.17 30.9

Includes the District of Columbia (DC). Also shown are population, commercial building roof area, commercial building conditioned
roof area, and commercial sector prices of electricity and natural gas. To obtain a lifetime rate of cooling energy saving, heating energy
penalty, or emission reduction, multiply the corresponding annual value by the expected life of the roof. The present value of lifetime
energy-cost saving is the product of the annual energy-cost saving and the present value multiplier selected from Table 1 or computed
from Eq. 3

Acknowledgments This study was funded by the Sika
Sarnafil corporation. The writing of this document was
supported by the Assistant Secretary for Renewable Energy
under Contract No. DE-AC02-05CH11231. We would like to
thank Kevin Foley, Stanley Graveline, and Brian Whelan of
Sika Sarnafil for their comments. We would also like to thank
Art Diem of the US EPA for assistance with eGRID electricity
emission factors, and Miguel Echavarri of Intelimap, Inc. for
supplying eGRID mapping data.
Appendix
In a 1997 study, Konopacki et al. (1997) estimated
that immediately increasing the solar reflectance of all
roofs on a subset of US commercial buildings by 0.45
would incur an annual cooling energy saving of
2.9 TWh, an annual heating-energy penalty of 71
million therms, and an annual energy cost saving of
$210 million. To match the basic assumptions in the
current study—i.e., only 80% of commercial building
CRA retrofitted; roof solar reflectance increased by
0.35, rather than by 0.45—we can revise the 1997
national sums by a factor of 0:80
0:35 ¼ 0:62.
However, the national cooling energy saving, heating
energy penalty, and energy cost saving predicted by
the current study (10.4 TWh, 133 million therms, and
$735 million) are 6.0, 3.0, and 5.6 times greater,
respectively, than revised estimates from the 1997
study (1.8 TWh, 44 million therms, and $131
million). Several factors make the current national
sums exceed those computed in 1997.
1. When computing CRA from the 1992 Commer-
cial Buildings Energy Consumption Survey (EIA
1994b), the 1997 study included only seven
classes of building (office, retail store, primary
school, secondary school, hospital, nursing home,
and grocery store) out of the 21 classes surveyed.
We estimate from our analysis of the 2003
CBECS dataset that these seven building classes
(principle building activities) comprised only
38% of the total commercial building CRA.
Hence, the 1997 national sums should be multi-
plied by a factor of 1=0:38 ¼ 2:6 to account for
the omitted building classes.
2. The 2003 CBECS detailed tables (EIA 2006c)
report the total floor space of buildings with at
least some air conditioning at 5,900 million
square meters, which is 11% higher than the
value of 5,300 million square meters reported by
the 1992 CBECS detailed tables (EIA 1994c).
Therefore, the 1997 national sums should be
multiplied by factor of 1.1 to account for net
new construction.
3. For equal increases in roof solar reflectance
(0.35), the current study predicts greater cooling
energy savings and smaller heating energy penal-
108 Energy Efficiency (2010) 3:53–109

ties per unit CRA than does the 1997 study. For References
example, for a new office building in Atlanta,
GA, the current study predicts a per-CRA cooling Akbari, H., & Konopacki, S. (2005). Calculating energy-saving
energy saving 88% higher and a per-CRA heating potentials of heat-island reduction strategies. Energy
energy penalty 13% lower than the corresponding Policy, 33, 721–756.
1997 values (pre-scaled by a factor of 0.35/0.45= Akbari, H., Eto, J., Turiel, I., Heinemeier, K., Lebot, B.,
Nordman, B., & Rainer, L. (1989). Integrated estimation
0.78). We attribute most of the difference to the of commercial sector end-use load shapes and energy use
use in the current study of an improved roof intensities (Report No. LBL-27512). Berkeley, CA: Law-
assembly heat transfer module in DOE-2.1E. This rence Berkeley National Laboratory.
module incorporates radiative heat transfer, an Akbari, H., Rainer, L., & Eto, J. (1991). Integrated estimation
of commercial sector end-use load shapes and energy use
improved external convection coefficient, and intensities, Phase II (Report No. LBL-30401). Berkeley,
temperature-dependent thermal conductance of CA: Lawrence Berkeley National Laboratory.
insulation (Gartland et al. 1996). Hence, to Akbari, H., Eto, J., Konopacki, S., Afzal, A., Rainer, L., &
account for improvements in simulation of roof Heinemeier, K. (1993). Integrated estimation of commer-
cial sector end-use load shapes and energy use intensities
heat transfer, the 1997 national cooling energy in the PG&E service area (Report No. LBL-34263).
saving should be multiplied by a factor of 1.9, Berkeley, CA: Lawrence Berkeley National Laboratory.
and the 1997 national heating energy penalty Akbari, H., Konopacki, S., Parker, D., Wilcox, B., Eley, C., & Van
should be multiplied by a factor of 0.9. Geem, M. (1998). Calculations in support of SSP90.1 for
reflective roofs. ASHRAE Transactions, 104(1B), 984–995.
4. The year 2005 US average commercial sector Akbari, H., Konopacki, S., & Pomerantz, M. (1999). Cooling-
electricity and natural gas prices ($0.0867/kWh energy savings potential of reflective roofs for residential
and $1.12/therm) are 12% and 121% higher, and commercial buildings in the United States. Energy, 24,
respectively, than the corresponding 1993 values 391–407.
California Energy Commission (CEC). (1994). Technology
($0.0774/kWh and $0.506/therm; EIA 2007b, c). energy savings volume II: Building prototypes (Report
No. P300-94-007). Sacramento, CA: Author.
Adjusting the 1997 estimate of US annual cooling DOE-2. (2007). Lawrence Berkeley National Lab DOE-2
energy saving by the factors listed above yields website. Available from http://simulationresearch.lbl.gov/
dirsoft/d2whatis.html.
2:9 TWh
×0:80
2.9 TWh 0.80ðroof
(rooffraction
fractionupgradedÞ
upraded)
0:35=0:45
× 0.35/0.45 Energy Information Administration (EIA) (1979). Commercial
ðsmaller increase in
(smaller increase in solar reflectance)

solar reflectanceÞ 2.6 ðmore
× 2:6 (more buildings consumption and expenditures (nonresidential
building classesÞ
1:1 ðconstructionÞ
1:9 ðbetter
building classes) × 1.1 (construction) × 1.9 (better sim buildings energy consumption survey). Washington, DC:
ulationÞ ¼ 9:8 TWh, which is 6%
simulation) = 9.8 TWh, which is 6% below thebelow the current US Department of Energy.
Energy Information Administration (EIA) (1983). Commercial
study’s estimate
estimate of
of 10.4
10.4TWh.
TWh.
buildings consumption and expenditures (nonresidential
Adjusting the 1997 estimate of US annual heating buildings energy consumption survey). Washington, DC:
energy penalty by the the factors
factors listed
listed above
above yields
yields 133
133 US Department of Energy.
million therms
× 0:80
million therms 0.80ðroof
(roof fraction upgraded)
×0:35=
fractionupgradedÞ 0.35/ Energy Information Administration (EIA) (1994a). Commercial
buildings consumption and expenditures. Washington,
0.45 ðsmaller
0:45 (smaller increase
increasein insolar reflectance)
×2:6
solarreflectanceÞ 2.6ðmore
(more DC: US Department of Energy.
building classes)
×
buildingclassesÞ ðconstructionÞ
1:11.1
0:9×ðbetter
(construction) sim
0:9 (better Energy Information Administration (EIA). (1994b). 1992
ulationÞ ¼ 114
simulation) million
= 114 million therms,
therms,which
whichisis16%
16% below commercial buildings energy consumption survey public use
the current study’s estimate of of 133
133 million
million therms.
therms. microdata [Data set]. US Department of Energy. Retrieved
from http://www.eia.doe.gov/emeu/cbecs/92microdat.html.
Computing US annual energy cost saving from
Energy Information Administration (EIA). (1994c). 1992 commer-
2005 US-average commercial sector energy prices cial buildings energy consumption survey detailed tables
and the adjusted values of cooling energy saving and [Data set]. US Department of Energy. Retrieved from http://
heating energy penalty yields 9:8 TWh
$0:0867= www.eia.doe.gov/emeu/cbecs/detailed_tables_1992.html.
Energy Information Administration (EIA). (2006a). Electric
kWh − 114 million therms
$1:12=therm ¼ $722M,
power annual 2005, Table 4 [Data set]. US Department of
which is 2% lower than the current study’s estimate of Energy. Retrieved from http://www.eia.doe.gov/cneaf/elec
$735 million. tricity/esr/table4.xls
Hence, the adjusted 1997 estimates of national Energy Information Administration (EIA). (2006b). 2003
commercial buildings energy consumption survey
annual cooling energy saving, heating energy penalty,
(CBECS) public use microdata [Data set]. US Department
and energy cost saving agree quite well with those of Energy. Retrieved from http://www.eia.doe.gov/emeu/
made in the current study. cbecs/cbecs2003/public_use_2003/cbecs_pudata2003.html.
Energy Efficiency (2010) 3:53–109
Energy Information Administration (EIA). (2006c). 2003
commercial buildings energy consumption survey (CBECS)
detailed tables [Data set]. US Department of Energy.
Retrieved from http://www.eia.doe.gov/emeu/cbecs/
cbecs2003/detailed_tables_2003/detailed_tables_2003.html.
Energy Information Administration (EIA). (2007a). Natural gas
commercial prices, 1967–2006 [Data set]. US Department
of Energy. Retrieved from http://tonto.eia.doe.gov/dnav/
ng/xls/ng_pri_sum_a_EPG0_PCS_DMcf_a.xls.
Energy Information Administration (EIA). (2007b). Current and
historical monthly retail sales, revenues, and average retail
price by state and by sector (Form EIA-826) [Data set]. US
Department of Energy. Retrieved from http://www.eia.doe.
gov/cneaf/electricity/page/sales_revenue.xls.
Energy Information Administration (EIA). (2007c). U.S. natural gas
summary. US Department of Energy. Retrieved from http://
tonto.eia.doe.gov/dnav/ng/xls/ng_sum_lsum_dcu_nus_a.xls.
Environmental Protection Agency (EPA). (2000). Emission facts:
Average annual emissions and fuel consumption for passenger
cars and light trucks (Document No. EPA420-F-00-013).
Retrieved from http://www.epa.gov/otaq/consumer/f00013.pdf.
Environmental Protection Agency (EPA). (2005). Compilation
of air pollutant emission factors, volume I: Stationary
point and area source (AP 52, 5th edn). Retrieved from
http://www.epa.gov/ttn/chief/ap42.
Environmental Protection Agency (EPA). (2007). Emissions &
generation resource grid integrated database (eGRID)
2006 v2.1 [Data set]. Retrieved from http://www.epa.gov/
cleanenergy/egrid.
ESRI. (2007a). ESRI ArcGIS Desktop 9.1 [Software]. Avail-
able from http://esri.com.
ESRI. (2007b). ESRI Data & Maps 2005 [Data set]. Available
from http://esri.com.
109
Gartland, L. M., Konopacki, S. J., & Akbari, H. (1996).
Modeling the effects of reflective roofing. In Proceed-
ings of the 1996 ACEEE Summer Study on Energy
Efficiency in Buildings. Pacific Grove, CA: American
Council for an Energy-Efficient Economy (vol. 4, pp.
117–124).
Konopacki, S., & Akbari, H. (1998). Simulated impact of roof
surface solar absorptance, attic, and duct insulation on
cooling and heating energy use in single-family new
residential buildings (Report No. LBNL-41834).
Berkeley, CA: Lawrence Berkeley National Laboratory.
Konopacki, S., Akbari, H., Pomerantz, M., Gabersek, S., &
Gartland, L. (1997). Cooling energy savings potential of
light-colored roofs for residential and commercial build-
ings in 11 US metropolitan areas (Report No. LBNL-
39433). Berkeley, CA: Lawrence Berkeley National
Laboratory.
Konopacki, S., Gartland, L., Akbari, H., & Rainer, L. (1998).
Demonstration of energy savings of cool roofs (Report No.
LBNL-40673). Berkeley, CA: Lawrence Berkeley Nation-
al Laboratory.
Levinson, R., Akbari, H., & Gartland, L. M. (1996). Impact of
the temperature dependency of fiberglass insulation R-
value on cooling energy use in buildings. In Proceedings
of the 1996 ACEEE Summer Study on Energy Efficiency
in Buildings. Pacific Grove, CA: American Council for an
Energy-Efficient Economy (vol. 10, pp. 85–94).
Levinson, R., Akbari, H., Konopacki, S., & Bretz, S. (2005).
Inclusion of cool roofs in nonresidential Title 24 prescrip-
tive requirements. Energy Policy, 33(2), 151–170.
National Renewable Energy Laboratory (NREL). (2007).
TMY2s typical meteorological years. Retrieved from
http://rredc.nrel.gov/solar/pubs/tmy2.