Updates on Revision to ASHRAE Standard 90.

2:
Including Roof Reflectivity for Residential Buildings

H. Akbari, Lawrence Berkeley National Laboratory

S. Konopacki, Lawrence Berkeley National Laboratory
D. Parker, Florida Solar Energy Center

ABSTRACT

This paper discusses the results of a simulation effort in support of ASHRAE SSPC
90.2 for inclusion of reflective roofs in the proposed standard. Simulation results include
the annual electricity and fuel use for a prototypical single-family one-story house. In
order to maintain consistency with the other requirements of the draft standards, we used
the 90.2 Envelope Sub-committee DOE-2 prototype building and operating schedules
which were supplied to us. The parametric simulations were performed for the following
scenarios and combinations thereof: 3 heating systems, 4 duct and duct insulation
configurations, 5 levels of ceiling insulation, 4 levels of roof reflectivity, and 4 levels of
attic air change rate. The simulations were performed for 32 climate regions.
The results are condensed into climate-dependent adjustment factors that give
equivalent reductions in roof insulation levels corresponding to increased roof
reflectivity. The equivalence is designed such that the net energy use (cooling plus heat-
ing) of the building stays constant when compared with energy use of a dark-colored
roof. Results indicate that in hot climates, increasing the roof reflectivity from 20% to
60% is worth over half of the roof insulation.

Introduction

Most commercial and residential buildings have dark roofs. Dark roofs are heated
by the summer sun and this raises the summertime cooling demand. For highly absorp-
tive (low-solar reflectance) roofs, the difference between the surface and ambient air tem-
peratures may be as high as 50°C (90°F), while for less absorptive (high-solar
reflectance) roofs, such as white paint, the difference is only about 10°C (18°F). For this
reason, "cool" roofs (which absorb little "insolation") are effective in reducing cooling
energy use. Numerous experiments in several residential and small commercial build-
ings in California and Florida show that painting roofs white reduces air-conditioning
energy use (compressor and condenser unit) between 10 and 50% (ranging from $10 to
$100 per year per 100m2), depending on the amount of thermal resistance of insulation
under the roof (Akbari et al. 1997, Parker et al. 1998). The savings, of course, are strong
functions of the thermal integrity of a building and climate conditions.
The American Society of Heating, Refrigeration, and Air-conditioning Engineers
(ASHRAE) develops voluntary standards to improve energy efficiency in buildings. In
many applications, the voluntary ASHRAE standards are modified by states, federal, and
other governmental organizations and used as codes and standards. Two such standards
address energy efficiency in new buildings: ASHRAE Standard 90.1 (Standards for
Buildings Except Low-Rise Residential Buildings) and ASHRAE Standard 90.2
(Energy-Efficient Design of New Low-Rise Residential Buildings). In 1998, Standard
90.1 adopted modification to the existing standards (Akbari et al. 1998, ASHRAE 1999).
Prior to adoption of the standards for inclusion of reflective roofs, ASHRAE sponsored a
symposium to discuss present results from field application and modeling (ASHRAE
1998). The Envelope Subcommittee of ASHRAE Standard 90.2 also recognized the
importance of roof reflectivity in residential buildings in reducing the net energy con-
sumption of a given building, and it organized a task group to develop a proposal to
modify the existing standards. In order to be consistent with other sections of the pro-
posed standards, the task group planned a detailed simulation approach to study the
impact of reflective roofs on heating and cooling energy use of of several prototypical
buildings over a wide range of climates. This paper summarizes the results of the simu-
lation effort in support of ASHRAE SSPC 90.2 for inclusion of reflective roofs in the
proposed standard.

Methodology

Reflective roofs reduce the flow of heat into the building by reflecting most of the
incident solar radiation during hot summer days. Having a well-insulated roof will also
reduce the heat gains during the day. During those hours of the day when the ambient
temperature is lower than the inside temperature, having high insulation in the roof
would block the path of heat flow out of the building. During the winter when the days
are short and cloudy and the sun angle is low, a reflective roof may add a heating penalty.
Therefore, we analyzed the impact of the roof reflectance in terms of a trade-off with roof
insulation. On that basis, the Envelope Subcommittee directed us to perform comprehen-
sive simulations to analyze cooling energy savings and heating energy penalties of
several prototypical buildings over a wide spectrum of climatic conditions. The DOE-
2.1E building energy simulation program was selected as the tool to perform this
analysis.
We used a residential building prototype that ASHRAE has used extensively in
support of developing criteria for Standard 90.2. The details of the prototypical building
are summarized in Table 1. The building was simulated with electric cooling, electric
heat pump, electric resistance heating, and gas heating systems.
Our simulations included prototypes with and without attics. These building were
simulated for a variety of roof insulation and roof reflectances. The roof insulations
included ceiling insulation levels: R-1, R-11, R-19, R-30, R-49. Parametric for roof
reflectivity included reflectance of 0.10, 0.25, 0.50, and 0.75. In addition we modeled
distribution system configurations with ducts in the attics with three levels of duct insula-
tion (R-2, R-4, and R-6) and ducts in the conditioned space. For the prototypical build-
ings with an attic, a fractional leakage area of 1:300 for the attic was assumed.
The simulations were performed for a wide range of climatic conditions from very
hot to very cold. A total of 36 climates were considered for these simulations; weather
data for five of these locations were not available. Also, for the Los Angeles area, simu-
lations were performed for both LAX, and Long Beach. Hence, in total, the simulations
were performed for 32 climates. These climate conditions are shown in Table 2.
The locations of the distribution ducts have a significant impact on the energy per-
formance of cooling systems. Leaky ducts in attics with a low level of duct insulation
can significantly reduce the efficiency of duct systems. Jump and Modera (1994) have
measured the duct efficiency and reported a reduced efficiency of as much as 50% in
some residences. The higher the temperature of the attic, the higher the inefficiencies of
the duct systems. Parker et al. (1998) have developed a model to account for the impact
of attic temperature on the performance of the cooling systems. In our simulations, we
augmented DOE-2 with the algorithm developed by Parker et al.
Upon completion of simulated heating and cooling energy use, we regressed the
results into quadratic functions of roof absorptance (1 - reflectance), α, and u-value, U, of
the roof system. The equation used is:
Ei = C0 + C1U + C2U2 + C3Uα (1)
Where, Ei is either annual electricity use in kWh, annual gas energy use in therms, or net
energy use in $. To obtain the net energy-use cost, we used the 1998 national average of
$0.0826/kWh and $0.691/therm for the price of electricity and gas, respectively (EIA
1998). This linear correlation proved to be adequate for our analysis; the 95%
confidence accuracy is 2%, the 98% accuracy is 3%.
We used these correlations to estimate the equivalency of the u-values and roof
absorptance. That is: given the energy use of a building with a dark roof (high absorp-
tance = α1) and an overall u-value of U1, what will be the new overall u-value (U2) if the
roof had a higher reflectivity (α2 < α1), such that the annual energy use remains the
same? Applying this equivalency condition, the level of roof insulation requirements in
most hot cities could be reduced by a factor of 2.
To optimize the energy use of the building, Akbari et al. (1998) recommended
using a square root correlation of
U2 U2 1/2
( hhh ) = ( hhh ) (2)
U1 Recom U1 Equivalent
where, Recom is the recommended value and Equivalent is the equivalency of the roof
absorptance and u-value obtained from the correlations.

Results

Table 3 shows the simulated annual energy expenditure for three climate regions:
Phoenix (hot and dry), Sacramento (moderate and dry), and Madison (cold). The results
are shown for heat pump, electric resistance, and gas heating systems. In Phoenix, for gas
heating systems, savings in the range of 6-17% (for various level of roof insulation) are
estimated by increasing the roof reflectance from 0.10 to 0.50, In Sacramento, the sav-
ings are in the range of 4-11%. Savings for Madison are nil. Since the price of gas per
unit of delivered energy is smaller than that of electricity, the savings are smaller for
electric heat pump and resistance heating.
The impact of roof reflectivity on the required level of roof insulation is shown in
Table 4. In hot climates, a significant amount of roof insulation can be saved by increas-
ing the roof reflectivity. For example in Phoenix, a roof system with a reflectivity of
10% and ceiling insulation of R-30 has an equivalent annual energy performance of a
roof system with a reflectivity of 50% and ceiling insulation of R-14; over 50% savings
in required R-value of the insulation. Lower levels of insulation savings are observed in
moderate climates such as Sacramento. In cold climates, the saving in roof insulation is
obviously nil.
We performed a detailed sensitivity analysis in looking at the impact of variation
in duct R-value and attic leakage area fractions on the overall U2/U1. In general, in most
cases the impact was smaller than 10%. Hence, for the reminder of the analysis for the
prototypes with an attic, we assumed a duct insulation of R-4 and a leakage area fraction
of 1:300.
The results of the analysis for all climate regions are shown in Table 5, in an
ascending ratio of heating-degree-days (base 65F) over cooling-degree-days (base 65).
Our recommended U2/U1 values are significantly lower than those obtained from the
correlations. Finally, we grouped the results into bins of similar modification based on
the heating-degree-days.

ASHRAE Proposal

Based on this analysis, the ASHRAE Envelope Subcommittee of Standard 90.2

voted unanimously to adopt the following proposal as a modification to SSP 90.2.

"Section 5.3.1.1: Single-Family Buildings (Ceiling with attics)

Exception to 5.3.1.1: For roofs where the exterior surface has either: a) a minimum total
solar reflectance of 0.65 when tested in accordance with ASTM E903 or E1918, and has a
minimum thermal emittance of 0.75 when tested in accordance with ASTM E408 or C1371;
or b) has a minimum solar reflectance index (SRI) of 75 calculated in accordance with
ASTM E1980 for medium wind-speed conditions, the U-value of the proposed ceiling shall
be permitted to be adjusted using Equation 5-3.1 for demonstrating compliance:

Uceiling_adj = Uceiling_proposed X Multiplier (5-3.1)

Where:
Uceiling_adj = the adjusted ceiling U-value for use in demonstrating compliance
Uceiling_proposed = the U-value of the proposed ceiling, as designed
Multiplier = the ceiling U-value multiplier from Table 5.3.1.

Section 5.3.1.2: Single-Family Buildings (Ceilings without attics)

Exception to 5.3.1.2: For roofs where the exterior surface has either: c) a minimum total
solar reflectance of 0.65 when tested in accordance with ASTM E903 or E1918, and has a
minimum thermal emittance of 0.75 when tested in accordance with ASTM E408 or C1371;
or d) has a minimum solar reflectance index (SRI) of 75 calculated in accordance with
ASTM E1980 for medium wind-speed conditions, the U-value of the proposed ceiling shall
be permitted to be adjusted using Equation 5-3.1 for demonstrating compliance:

Uceiling_adj = Uceiling_proposed X Multiplier (5-3.1)

Where:
Uceiling_adj = the adjusted ceiling U-value for use in demonstrating compliance
Uceiling_proposed = the U-value of the proposed ceiling, as designed
Multiplier = the ceiling U-value multiplier from Table 5.3.1.

Section 5.5.1.1: Multi-Family Buildings (Ceilings with attics)

Exception to 5.5.1.1: For roofs where the exterior surface has either: e) a minimum total
solar reflectance of 0.65 when tested in accordance with ASTM E903 or E1918, and has a
minimum thermal emittance of 0.75 when tested in accordance with ASTM E408 or C1371;
 or f) has a minimum solar reflectance index (SRI) of 75 calculated in accordance with
ASTM E1980 for medium wind-speed conditions, the U-value of the proposed ceiling shall
be permitted to be adjusted using Equation 5-3.1 for demonstrating compliance:

Uceiling_adj = Uceiling_proposed X Multiplier (5-3.1)

Where:
Uceiling_adj = the adjusted ceiling U-value for use in demonstrating compliance
Uceiling_proposed = the U-value of the proposed ceiling, as designed
Multiplier = the ceiling U-value multiplier from Table 5.3.1.

Section 5.5.1.2: Multi-Family Buildings (Ceilings without attics)

Exception to 5.5.1.2: For roofs where the exterior surface has either: g) a minimum total
solar reflectance of 0.65 when tested in accordance with ASTM E903 or E1918, and has a
minimum thermal emittance of 0.75 when tested in accordance with ASTM E408 or C1371;
or h) has a minimum solar reflectance index (SRI) of 75 calculated in accordance with
ASTM E1980 for medium wind-speed conditions, the U-value of the proposed ceiling shall
be permitted to be adjusted using Equation 5-3.1 for demonstrating compliance:

Uceiling_adj = Uceiling_proposed X Multiplier (5-3.1)

Where:
Uceiling_adj = the adjusted ceiling U-value for use in demonstrating compliance
Uceiling_proposed = the U-value of the proposed ceiling, as designed
Multiplier = the ceiling U-value multiplier from Table 5.3.1.

Table 5.3.1. Ceiling U-value Multiplier

iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii
c HDD 65 (HDD18) Ceilings with Attics Ceilings without Attics c
c 0-360 (0-200) 1.50 1.30 c
c 361-900 (201-500) 1.30 1.30 c
c 901-1800 (501-1000) 1.20 1.30 c
c 1801-2700 (1001-1500) 1.15 1.30 c
c c
c 2701-3600 (1501-2000) 1.10 1.20 c
ciiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii
> 3600 (> 2000) 1.00 1.00 c

Section 8.8.3.1: Exterior Absorptivity

Since the colors are subject to change over the life of the building, the exterior absorptivity
of all walls and roofs shall be 0.5 regardless of color, and the exterior absorptivity of roofs
shall be 0.2 regardless of color. If unconditioned spaces so as garages are not modeled,
walls between them and conditioned space shall be treated as exterior walls with an absorp-
tivity of zero.
Note: For low absorptivity roofs, the reference house may employ Exceptions 5.3.1.1 or
5.3.1.2 or 5.5.1.1 or 5.5.1.2."

Conclusion

In this study, we have documented the result of a building energy simulation

analysis to account for energy-saving benefits of reflective roofs in residential buildings.
DOE-2 was used to calculate the annual electricity and fuel use for a prototypical single-
family one-story house in 32 climate regions. Parametric simulations were performed for
the following scenarios and combinations thereof: 3 heating systems, 4 duct and duct
insulation configurations, 5 levels of ceiling insulation, 4 levels of roof reflectivity, and 4
levels of attic air change rate.
The results are condensed into climate-dependent adjustment factors that give
equivalent reductions in roof insulation levels corresponding to increased roof
reflectivity. The equivalence is designed such that the net energy use (cooling plus heat-
ing) of the building remains constant when compared with energy use of a dark-colored
roof. Results indicate that in hot climates, by increasing the roof reflectivity from 20% to
60%, one can reduce the roof insulation by half and still have the same net annual energy
use.

Acknowledgement

This work was supported by the U.S. Environmental Protection Agency (EPA)
under IAG No. DW89938442-01-2 and by the Assistant Secretary for Energy Efficiency
and Renewable Energy, Building Technologies, of the U.S. Department of Energy (DOE)
under contract No. DE-AC03-76SF00098.

References
ASHRAE. 1999. "ASHRAE/IESNA Standard 90.1-1999: Energy Standard for Buildings
Except Low-Rise Residential Buildings," Page 20, American Society of Heating,
Refrigerating and Air Conditioning Engineers 1791 Tullie Circle, NE, Atlanta, Geo.
30329.
ASHRAE, 1998. "ASHRAE Technical Bulletin, Energy Savings of Reflective Roofs,"
American Society of Heating Refrigerating and Air-Conditioning Engineers,
Atlanta, Geo., Volume 14, Number 2, January.
Akbari, H., S. Konopacki, D. Parker, B. Wilcox, C. Eley, and M. Van Geem. "Calcula-
tions in Support of SSP90.1 for Reflective Roofs," ASHRAE Transactions,, 104(1),
pp. 984-995, January 1998.
Akbari, H., S. Bretz, H. Taha, D. Kurn, and J. Hanford. 1997. "Peak Power and Cooling
Energy Savings of High-albedo Roofs," Energy and Buildings, Vol. 25, No. 2, pp.
117-126.
EIA 1998. Energy Information Administration (EIA). 1998. http://www.eia.doe.gov.
Washington, DC.
Jump, D. and M. Modera. 1994. "Energy Impacts of Attic Duct Retrofits in Sacramento
Houses," Lawrence Berkeley National Laboratory Report LBNL-35375, Berkeley,
CA.
Parker, D., J. Huang, S. Konopacki, L. Gartland, J. Sherwin and L. Gu. 1998. "Measured
and Simulated Performance of Reflective Roofing Systems in Residential Build-
ings". ASHRAE Transactions 104(1):963-975.
Table 1. Prototypical construction, equipment, and operation characteristics for a single-
family one-story ranch house.
i iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii
c Construction c
c c
c zones interior: conditioned c
c attic: unconditioned and naturally ventilated c
c floor area 1540ft2 c
c perimeter 166ft c
c c
c aspect ratio 1:1 c
c wall height 8ft c
c roof 1/4" asphalt shingle over 3/4" plywood decking (4/12 slope) c
c solar absorptance: 0.90, 0.75, 0.50, or 0.25 c
c c
c infrared emittance: 0.9 c
c overhang 2ft around entire perimeter c
c ceiling frame (15%) and R-1, 11, 19, 30, or 49 fiberglass insulation (85%) c
c c
c over 1/2" drywall c
c exterior wall stucco over frame (15%) and R-11 fiberglass insulation over 1/2" drywall c
c windows 185ft2 (14% of exterior wall area) double clear with operable shades, U- c
c IP 0.57, and shading coefficient 0.88 c
c c
c foundation slab-on-grade with carpet and pad c
c Equipment c
c sizing based on peak cooling and heating loads c
c sizing ratio 1.25 c
c c
c cooling direct expansion: SEER 10 c
c heating (1) gas furnace: AFUE 78% c
c (2) electric heat pump: HSPF 6.8 c
c c
c (3) electric resistance c
c distribution constant-volume forced air system c
c 10% duct leakage c
c duct insulation R-value: 2, 4, 6 (attic), or 0 (interior) c
c 2 c
c supply duct area 370ft c
c return duct area 69ft2 c
c Operation c
c cooling thermostat 78°F c
c c
c heating thermostat 68°F c
c natural ventilation enthalphic controlled window operation: 68°F min and 5 ACH max c
c infiltration Sherman-Grimsrud: fla 1:2000 (interior) and fla 1:300 (attic) c
c 2 c
peak internal heat gain 0.68 W/ft
ci iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii c
Table 2. Selected locations, TMY2 weather file availability and degree-days.
iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii
c id
iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii location tmy2 weather file c c cdd 65 hdd 65 c
c cc c
c 1 Adak, AK not available cc c
c 2 Albuquerque, NM Albuquerque, NM cc 1211 4361 c
c 3 Brownsville, TX Brownsville, TX cc 3563 659 c
c 4 Bangor, ME not available c c c
c 5 Bismarck, ND Bismarck, ND c c 408 8666 c
c 6 Bryce, UT not available c c c
c cc c
c 7 Charleston, SC Charleston, SC cc 2010 2209 c
c 8 Denver, CO Boulder, CO cc 623 6007 c
c 9 Dodge, KS Dodge City, KS cc 1371 5353 c
c 10 El Paso, TX El Paso, TX c c 2046 2597 c
c 11 Fort Worth, TX Fort Worth, TX c c 2415 2304 c
c 12 Fairbanks, AK Fairbanks, AK c c 29 14095 c
c cc c
c 13 Fresno, CA Fresno, CA cc 1884 2602 c
c 14 Fort Smith, AR Fort Smith, AR cc 1895 3351 c
c 15 Honolulu, HI Honolulu, HI cc 4329 0 c
c 16 Jacksonville, FL Jacksonville, FL c c 2657 1437 c
c 17 Kwajalein, PI St.Paul Island, AK c c 0 11126 c
c 18 Lake Charles, LA Lake Charles, LA c c 2624 1683 c
c cc c
c 19 Laredo, TX not available cc c
c 20 Las Vegas, NV Las Vegas, NV cc 3067 2293 c
c 21 Los Angeles, CA LAX cc 470 1291 c
c 21 Los Angeles, CA Long Beach c c 943 1309 c
c 22 Miami, FL Miami, FL c c 4127 141 c
c cc
23 Madison, WI Madison, WI 521 7495 c
c cc c
c 24 North Omaha, NE Omaha, NE cc 1051 6047 c
c 25 New York, NY New York, NY cc 1002 5090 c
c 26 Phoenix, AZ Phoenix, AZ cc 3815 1154 c
c 27 Redmond, OR Redmond, OR c c 194 6732 c
c 28 Roswell, NM not available c c c
c cc
29 Tucson, AZ Tucson, AZ 2763 1554 c
c cc c
c 30 Sacramento, CA Sacramento, CA cc 1144 2794 c
c 31 San Diego, CA San Diego, CA cc 766 1076 c
c 32 Seattle, WA Seattle, WA cc 127 4867 c
c 33 San Francisco, CA San Francisco, CA c c 69 3239 c
c 34 St. Louis, MO St. Louis, MO c c 1437 5021 c
c cc
35 Washington, DC Sterling, VA 1044 5233 c
c cc c
c 36
iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii
Winnemucca, NV Winnemucca, NV cc 604 6444 c
Table 3. Simulated annual cooling and heating total energy base use [$/1000ft2 ] and the direct savings [%] from the use of high-
albedo roofs for a typical single-family one-story ranch house with gas heat, R-4 attic ducts and 1:300 attic fractional leakage area.

Phoenix Sacramento Madison

Ceiling insulation→ R-1 R-11 R-19 R-30 R-49 R-1 R-11 R-19 R-30 R-49 R-1 R-11 R-19 R-30 R-49
Gas furnace
base use, α=0.90 639 411 379 359 345 422 255 229 214 202 779 572 532 508 491
savings, α=0.75 6 4 3 3 2 4 2 2 2 1 0 0 0 0 0
savings, α=0.50 17 10 8 7 6 11 6 5 5 4 1 0 0 0 0
savings, α=0.25 28 17 14 12 10 17 10 8 7 7 1 -1 -1 -1 0
Electric heat pump
base use, α=0.90 687 444 409 388 372 525 346 316 298 283 1166 997 953 925 904
savings, α=0.75 6 3 3 3 2 3 1 1 1 1 0 0 0 0 0
savings, α=0.50 16 9 8 7 6 8 4 3 3 2 1 -1 -1 -1 -1
savings, α=0.25 26 15 13 11 10 12 6 5 5 4 1 -1 -1 -1 -1
Electric resistance
base use, α=0.90 804 513 468 442 423 856 557 504 473 449 1947 1522 1425 1367 1324
savings, α=0.75 5 3 2 2 2 1 1 0 0 0 0 0 0 0 0
savings, α=0.50 13 8 6 5 5 3 2 1 1 1 -1 -1 -1 -1 -1
savings, α=0.25 21 13 10 9 8 5 2 2 2 2 -2 -2 -2 -2 -1
Table 4. Estimated roof-composite U_factor (U2/U1) and equivalent cool-roof ceiling insulation R-value (shown in parentheses) as
a function of roof solar absorptance (α ) and ceiling insulation R-value. Dark roof absorptance is 0.90. All results for R-4 attic ducts
and 1:300 attic fractional leakage area.

Phoenix Sacramento Madison

ceiling insulation→ R-11 R-19 R-30 R-49 R-11 R-19 R-30 R-49 R-11 R-19 R-30 R-49
gas furnace
α = 0.75 1.14 1.16 1.17 1.18 1.07 1.08 1.08 1.08 1.00 1.00 1.00 1.00
(9) (15) (23) (37) (10) (17) (26) (42) (11) (19) (30) (49)
α = 0.50 1.48 1.54 1.59 1.64 1.21 1.23 1.24 1.25 1.01 1.01 1.01 1.01
(5) (9) (14) (21) (8) (14) (21) (33) (11) (19) (30) (48)
α = 0.25 1.97 2.16 2.33 2.50 1.39 1.42 1.45 1.47 1.01 1.01 1.01 1.01
(2) (5) (7) (10) (6) (11) (16) (25) (11) (19) (29) (48)
electric heat pump
α = 0.75 1.13 1.14 1.15 1.16 1.05 1.06 1.06 1.06 1.00 1.00 1.00 1.00
(9) (15) (24) (38) (10) (17) (27) (44) (11) (19) (30) (49)
α = 0.50 1.42 1.47 1.51 1.55 1.16 1.17 1.17 1.17 1.00 1.00 1.00 1.00
(6) (10) (15) (23) (8) (15) (23) (37) (11) (19) (30) (49)
α = 0.25 1.85 2.00 2.12 2.25 1.29 1.30 1.31 1.31 1.00 1.00 1.00 1.00
(3) (6) (8) (12) (7) (12) (19) (30) (11) (19) (30) (49)
electric resistance
α = 0.75 1.10 1.11 1.11 1.12 1.02 1.02 1.02 1.02 0.98 0.99 0.99 0.99
(9) (16) (25) (40) (11) (18) (29) (47) (11) (19) (31) (50)
α = 0.50 1.31 1.34 1.36 1.38 1.05 1.06 1.06 1.06 0.96 0.96 0.96 0.96
(7) (12) (18) (28) (10) (17) (27) (44) (12) (20) (32) (53)
α = 0.25 1.60 1.67 1.73 1.79 1.09 1.09 1.10 1.10 0.94 0.94 0.94 0.94
(4) (8) (12) (18) (9) (16) (26) (41) (12) (21) (33) (55)
Table 5. Estimated roof-composite U_factor (U2/U1) for a reflective roof with a solar absorptance of 0.45. Dark roof absorptance is
0.90. All results for R-4 attic ducts and 1:300 attic fractional leakage area.

Duct in Attics Ducts in Conditioned Space Roofs with no Attics

1/2 1/2
Location cdd_65 hdd_65 hdd/cdd U2/U1 (U2/U1) Recomm U2/U1 (U2/U1) Recomm U2/U1 (U2/U1)1/2 Recomm
Honolulu, HI 4329 0 0.00 2.62 1.62 1.5 2.62 1.62 1.5 1.69 1.30 1.3
Miami, FL 4127 141 0.03 2.19 1.48 1.5 2.19 1.48 1.5 1.67 1.29 1.3
Brownsville, TX 3563 659 0.18 1.6 1.26 1.25 1.6 1.26 1.25 1.61 1.27 1.3
Phoenix, AZ 3815 1154 0.30 1.53 1.24 1.2 1.53 1.24 1.2 2.39 1.55 1.3
Jacksonville, FL 2657 1437 0.54 1.52 1.23 1.2 1.52 1.23 1.2 1.59 1.26 1.3
Tucson, AZ 2763 1554 0.56 1.49 1.22 1.2 1.49 1.22 1.2 2.76 1.66 1.3
Lake Charles, LA 2624 1683 0.64 1.46 1.21 1.2 1.46 1.21 1.2 2.02 1.42 1.3
El Paso, TX 2046 2597 1.27 1.42 1.19 1.15 1.42 1.19 1.15 1.68 1.30 1.3
Los Angeles, CA 943 1309 1.39 1.38 1.17 1.2 1.38 1.17 1.2 1.64 1.28 1.3
San Diego, CA 766 1076 1.40 1.37 1.17 1.2 1.37 1.17 1.2 1.69 1.30 1.3
Las Vegas, NV 3067 2293 0.75 1.37 1.17 1.15 1.37 1.17 1.15 1.65 1.28 1.3
Fresno, CA 1884 2602 1.38 1.34 1.16 1.15 1.34 1.16 1.15 1.56 1.25 1.3
Charleston, SC 2010 2209 1.10 1.33 1.15 1.15 1.33 1.15 1.15 1.58 1.26 1.3
Fort Worth, TX 2415 2304 0.95 1.31 1.14 1.15 1.31 1.14 1.15 1.64 1.28 1.3
Fort Smith, AR 1895 3351 1.77 1.24 1.11 1.1 1.24 1.11 1.1 1.63 1.28 1.3
Sacramento, CA 1144 2794 2.44 1.22 1.10 1.1 1.22 1.10 1.1 1.61 1.27 1.2
Albuquerque, NM 1211 4361 3.60 1.19 1.09 1 1.19 1.09 1 1.4 1.18 1.2
Los_Angeles, CA 470 1291 2.75 1.16 1.08 1.2 1.16 1.08 1.2 1.55 1.24 1.2
St. Louis, MO 1437 5021 3.49 1.11 1.05 1 1.11 1.05 1 1.5 1.22 1.2
Washington, DC 1044 5233 5.01 1.09 1.04 1 1.09 1.04 1 1.23 1.11 1
Dodge, KS 1371 5353 3.90 1.09 1.04 1 1.09 1.04 1 1.34 1.16 1
North Omaha, NE 1051 6047 5.75 1.08 1.04 1 1.08 1.04 1 1.27 1.13 1
Denver, CO 623 6007 9.64 1.06 1.03 1 1.06 1.03 1 1.33 1.15 1
Winnemucca, NV 604 6444 10.67 1.06 1.03 1 1.06 1.03 1 1.28 1.13 1
New York, NY 1002 5090 5.08 1.05 1.02 1 1.05 1.02 1 1.28 1.13 1
Bismarck, ND 408 8666 21.24 1.02 1.01 1 1.02 1.01 1 1.25 1.12 1
Redmond, OR 194 6732 34.70 1.01 1.00 1 1.01 1.00 1 1.26 1.12 1
Madison, WI 521 7495 14.39 1.01 1.00 1 1.01 1.00 1 1.18 1.09 1
Seattle, WA 127 4867 38.32 0.97 0.98 1 0.97 0.98 1 1.12 1.06 1
Fairbanks, AK 29 14095 486.03 0.97 0.98 1 0.97 0.98 1 1.09 1.04 1
Kwajalein, PI 0 11126 ∞ 0.95 0.97 1 0.95 0.97 1 0.93 0.96 1
San Francisco, CA 69 3239 46.94 0.94 0.97 1 0.94 0.97 1 0.98 0.99 1