Skip to main content Skip to footer

Solar Reflectivity Studies

May 15, 2012

S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 2 C L I F T A N D MO N T E S – A M O R O S • 6 7
8070 Park Lane, Suite 400, Dallas, TX 75231
Phone: 972-437-4200 • Fax: 972-437-4562
E-mail: and
Recent skyscraper designs have extensive exterior glass paneling that may cause hazardous
glare to neighboring buildings or nearby traffic. Examples of recent problem designs
include the Disney Concert Hall in Los Angeles, CA; The Vdara Hotel in Las Vegas, NV,
which was reported in local newspapers to produce a “death ray” due to intense solar reflections
from the concave curtain wall geometry; and the Nasher Sculpture Center in Dallas,
TX, whose skylight features—tailored to filter indirect daylight to the art galleries—are now
subject to direct solar reflections from a new curved tower with a metallic-coated glass
Curtain Wall Design and Consulting (CDC) has developed a method to advance the current
state of the art for solar reflectivity studies. Computational fluid dynamics (CFD) is
used to emit a large number of rays and trace their trajectories inside a computational
domain. The analysis allows accumulation of rays on discrete elements, thus compiling an
intensity value at the unique element location.
VICENTE MONTES-AMOROS is a structural and façade engineer specializing in the
design and engineering of building envelope systems, including natural stone, precast concrete
panels, unitized and stick-built curtain walls, etc. He earned a maser’s degree in
façade engineering at the University of Bath in the United Kingdom. Montes-Amoros has led
the Solar Reflectivity program at Curtain Wall Desgin and Consulting, Inc. (CDC) since its
creation and has continued its development to the present. He has published diverse articles
on various building envelope topics.
CHARLES CLIFT is senior principal and president of Curtain Wall Design and
Consulting, Inc. (CDC). Clift has worked at CDC in Dallas, TX, for 30 years, providing engineering
design for curtain wall and other exterior cladding systems. His tall-building experience
includes the Lotte World Tower in Seoul; Trump Tower, Chicago; Abraj Al Bait
Towers, Mecca; CITIC Plaza, Guangzhou; U.S. Bank Tower, Los Angeles; The Center, Hong
Kong; Emirates Tower, Dubai; and Bank of America Plaza, Atlanta. Clift was named as the
Dallas AIA Consultant of the Year in 2010.
6 8 • C L I F T A N D MO N T E S – A M O R O S S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 2
Historically, the study of light has captured
the interest of many scientists and
scholars who have contributed to a better
understanding of its characteristics, behavior,
and effects. The study of light is subjective
in some specific aspects, such as color
and glare, but very objective in aspects
such as direction and reflection. It has been
demonstrated that even though every type
of light originates from an energy source
such as the sun, an electric lamp, a lit candle,
etc., most of the light we see in the
physical world is the result of reflected light
(Tippens 1999).
Reflection is a physical phenomenon
that has been studied and classified using
two different theories. One of the theories
describes reflection using Maxwell’s undulatory
electromagnetic theory. It is simpler,
however, to describe reflection by using the
ray-tracing theory, which is the second theory.
The tracing theory treats the light as
rays and is generally known as geometrical
optics, which is based on Huygens’s
When a light ray travels in a medium
and finds an obstacle such as a glass surface,
part of the incident ray is reflected,
and the rest is transmitted to the other side
of the obstacle—in this case, glass. The
transmitted portion changes direction when
passing through the glass. This phenomenon
is called refraction and is characterized
by Snell’s Law. Refer to Figure 1 for the
graphical representation of the refraction
When glass is installed on building
façades, the refracted portion of the incident
light will penetrate to the building interior.
Depending on the light, glass characteristics,
and some other factors, the light
transmitted exhibits a different range of
phenomena, such as heat gain and UV
The type of reflection that this new
methodology investigates is the portion of
the light that is “bounced” from the glass
surface and returned to the medium. The
reflected light-directional behavior is
described by the reflection laws listed below
(Serway 1997):
• The incident angle is equal to the
reflected angle (see Figure 2).
• The incident ray, the reflected ray,
and the line perpendicular to the
surface (the normal) are located on
the same plane.
The reflection produced by glass and
other smooth and polished surfaces is
called specular reflection. The reflection
from an irregular or rough surface is called
diffuse reflection.
This new methodology focuses only on
the reflection from flat architectural glass,
which is specular; and in this paper, we will
refer to it as reflection from this point on. It
is important to mention that the light
reflected from nonplanar or prismatic surfaces
follows a different trajectory than the
one illustrated in Figure 2. This new
methodology deals only with planar surfaces.
Following the laws of reflection, light
emitted by the sun will be reflected on
exposed surfaces in the built environment,
such as: pavements, walls, roofs, etc. The
amount of light reflected and absorbed by a
building’s façade depends on the properties
of the façade materials and their position
relative to incident rays. In the built environment,
some of these reflections might
produce discomfort or a hazard, depending
on their intensity and direction.
The solar light reflection from a building
depends on the geographic location of the
project, the orientation, the climatology, the
treatment of its surfaces, and other factors.
This new methodology identifies reflections
from a building’s façade due to the use of
architectural flat glass and provides some
characterization as to potential for hazardous
glare conditions.
For any given project, the variation in
daylight duration throughout the different
seasons of the year can be observed, recorded,
and predicted. The most influential factor
in the daylight duration is the earth’s
polar axis’s natural tilt, which is equal to
23.4 degrees; if it was not for this, the difference
between daytime and nighttime
would not be as evident during the year.
Figure 3 shows the earth’s movement
around the sun during the entire calendar
year; note that the tilt angle remains constant
throughout the year.
As the earth moves around the sun
throughout the year, the location of the sun
with respect to the horizon changes every
day. And as the earth revolves around its
axis, the position of the sun with respect to
S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 2 C L I F T A N D MO N T E S – A M O R O S • 6 9
Figure 1 – Light refraction. Figure 2 – Reflection law.
Figure 3 – Earth movement around the
7 0 • C L I F T A N D MO N T E S – A M O R O S S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 2
any given point on earth will vary from
minute to minute. Another factor that
affects the sun’s position with respect to the
horizon during the year is the geographical
position of any given project. The further
away from the equator, the greater the difference
in altitude during the seasons; and
hence, the difference between daytime
duration throughout the year becomes
more evident.
In order to calculate the sun’s position
throughout the year, one of the first steps is
to determine the sun’s altitude, which is
also known as the solar elevation angle. To
do this, the angle of declination needs to be
calculated in order to take into consideration
the earth’s natural tilt. The angle of
declination is defined as the angle between
the equator and the ecliptic plane, as shown
in Figure 4.
The angle of declination can be calculated
with the following formula:
D = Declination
N = Day number
The angle of declination at the vernal
equinox (March 21) and at the autumnal
equinox (September 23) is equal to 0º. The
angle of declination at the summer solstice
(June 21) is equal to 23.4º, and the one at
the winter solstice (December 21) is equal to
-23.4º. These four major events mark the
seasonal changes.
By definition, “altitude” is the height an
object is above the horizon. The altitude of
the sun varies throughout the day, and it
reaches its maximum around noon (varies
during daylight savings schedules). As
shown on Figure 4, the altitude of the sun
will also change according the season and
the previously listed factors. The solar position
in the sky is called the solar elevation
angle or sun’s altitude and can be calculated
for every day of the year. This angle is
formed between the sun’s apparent disk- or
altitude-ring and the horizon.
To calculate the sun’s altitude, the following
formula can be used (Figure 5 shows
the altitude):
γ = Solar elevation angle (altitude)
H = Hour angle
D = Declination
L = Project’s latitude
As the earth rotates around its axis
every day, the location of the sun with
respect to any given point varies. This variation
is called solar azimuth (refer to Figure
5 for the graphical representation). By definition,
the solar azimuth angle is the horizontal
angle formed at the ground plane
between the sun and a reference location. It
can be found using the following equation:
Z = Solar azimuth angle
γ = Solar elevation angle (altitude)
D = Declination
L = Project’s latitude
Using the information calculated from
the equations above, a sun path diagram
can be plotted in order to visualize the sun’s
trajectory during the year. Figure 6 shows
an example of a sun path diagram.
Figure 6 helps in understanding how
the sun interacts with a given project at different
times throughout the calendar year.
Different hyperbolic blue lines and vertical
blue lines are shown along the project’s
south orientation for projects located in the
northern hemisphere. Each vertical line
represents an hour increase; noon is located
right at the center of the diagram. Each
vertical line east of noon represents one
hour’s decrease from noon. Each vertical
line west of noon represents one hour
increase from noon.
Hyperbolic lines represent months in
the calendar. Starting from the dark hyperbolic
on the bottom, which represents
December, it can be observed that the sun
rises later in the day than it does in July.
July is represented by the hyperbolic line
farthest away from December. Moving back
and away from December, each hyperbolic
line represents one-month increase until
the last hyperbolic line is reached. Once the
last hyperbolic line is reached (July), we
move back towards the first hyperbolic line
at a one-month increase per hyperbolic line
(Editor’s note: this is more visually tracked
in color.)
Identifying the sun’s location at any
given day at any given time can easily be
done using this kind of diagram.
Figure 4 – Angle of declination.
Figure 5 – Solar elevation angle and
solar azimuth angle.
S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 2 C L I F T A N D MO N T E S – A M O R O S • 7 1
Solar reflectivity is a common phenomenon,
and it is caused by the interaction
between the reflective materials on the
façade and the structures around it (Naai-
Jung and Yen-Shih, 2000). It can produce
discomfort, and it can even be a threat for air
traffic when the light is returned in the form
of glare. There are three different glare types:
• Direct glare
• Reflective glare
• Disability glare
Direct glare is a phenomenon originated
from light sources that cast luminance
directly into the eye’s visual cone. Reflective
glare occurs when light rays bounce off a
surface and cause a level of luminance to be
perceived from the angle of incidence of the
reflection. Disability glare is a level of
change of luminosity that significantly
reduces visibility of the observer. The first
two types of glare listed above can create a
discomfort effect that we call discomfort
glare. For solar reflectivity on buildings, it
has been observed that most of the cases
dealing with glare are related to discomfort
glare rather than disability glare (Naai-Jung
and Yen-Shih, 2001a).
Energy performance criteria influence
architectural design, encouraging use of
reflective glass to reduce penetration of
solar radiation into the building interior.
However, while a highly reflective glass efficiently
blocks solar heat gain, it causes a
significant impact on the neighboring environment
due to exterior reflections.
Typical clear glass
has an exterior reflectance
value of 9%, whereas coated
reflective glass exhibits an
exterior reflectance value of
approximately 20% to 40%.
The limits of solar reflectivity
evaluation are subjective
since they depend on
diverse factors and dynamic
circumstances and are variable
for every person. Some
of the factors that influence
people’s reflectivity perception
are age, eye pigmentation,
eye sensitivity (especially
for people who have undergone
any type of eye surgery
in which the pupil’s ability to rapidly adapt
to light contrasts has been affected), and
Currently, there are few approaches for
criteria limiting solar reflectivity. One comparative
benchmark addressing reflectivity
is published and enforced by the Sydney
City Council. It states that materials used
on the exterior of buildings can result in
undesirable glare for pedestrians and
motorists, limiting the reflectivity of these
materials to 20% of visible light. It also
states that glare can impose additional heat
load on other buildings. Unfortunately, with
the erection of increasingly complex building
shapes, this limit might not address
reflected-light concentrations. A complex
building shape clad with materials with a
reflective coefficient of 20% could still produce
undesirable glare.
A veiling luminance limit of 500 candelas
per square meter for the comfort of
motorists was suggested by Hassall (1991),
but this limit goes hand in hand with his
proposed approach for determining the
solar reflectivity. Hassall’s methodology is
based on the preparation of sun path diagrams
for every aspect on the development.
After this, a check-zone diagram needs to be
completed in order to determine the areas
influenced by the reflections. Once this is
done, the Holladay formula is used to calculate
the luminance intensity. This is often
graphically represented by a glare protractor.
The Hassall approach is limited since it
cannot determine the duration of time over
which reflections occur to the surroundings,
and it also ignores the effect of the
type of glazing (Rofail et al. 2004). This
approach also fails to address the limits of
solar reflectivity on neighboring buildings.
In addition to reflected glare coming
from the glass, the metal frame may also
reflect light that might act as a “blinking”
light while a transportation vehicle passes
by the building. This dynamic phenomenon
depends on the relative movement between
glare and a fast-moving vehicle. Such glare
could possibly distract or delay the
response to a sudden traffic situation
(Rofail et al., 2004). This effect is not simulated
by this study.
Given the subjectivity of individuals’
sensitivity to glare, the lack of an industrywide
accepted criteria, and the absence of
any precedence related to the limits of this
type of nuisance, a combination of the
approaches described in the section above
to determine the problematic areas and
time of day of occurrence was used. The
proposed design criterion for a threshold of
acceptable intensity is based on comparison
with common sources of light. The upper
limit compares light reflected from buildings
to direct sunlight. Depending on a
building’s shape, orientation, and other factors,
reflected light could be concentrated at
a given area, like a magnifying glass. These
cases could produce disability glare, which
could potentially affect visibility of people
anywhere within the building’s domain.
As a part of the proposed threshold, a
custom scale was developed with this new
methodology, which provides results as a
fractional value of the reflected light.
Ultimately, intensity of reflected light
depends only on the material’s reflectance
coefficient. Therefore, by taking the results
from this custom scale, together with the
material’s reflectance coefficient, we can
obtain the intensity of reflected light, or
glare. This scale provides the advantage of
results as a fractional value of the reflected
light, thereby giving flexibility to designers
to change the specified glass without having
to run all the models again.
If the reflectance coefficient of the glass
is known—10%, for example—and we obtain
from this custom scale a value of 1.0, it
means that 1.0 times 10% of the incident
light is returned in the form of glare. On the
other hand, if we get an output of 10.0 from
this custom scale, it means that 10.0 times
Figure 6 – Sun path diagram.
7 2 • C L I F T A N D MO N T E S – A M O R O S S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 2
10% (glass reflectance coefficient) of the incident
light is the “amount” of reflected light.
In this last example, it represents that it
could potentially be equal to direct sunlight.
It is suggested here, as a base threshold,
that the intensity of reflectivity measured
at an individual location be limited to
no more than one times the natural intensity
at the project site (Figure 7).
This threshold is reasonable because the
project’s areas are already receiving direct
sunlight; hence, setting such value as the
maximum intensity limit is conservative.
In today’s market, several computer
modeling software products exist for different
purposes such as rendering, imaging, special
effects, design, architecture, and lighting.
These programs use a wide variety of techniques
to estimate the luminosity intensity
for any given model or image. But the complexity
of the cases has increased, and the
models used some years ago were developed
to run under the radiosity method, which
requires more time to compute problems.
Recent advances in computational power
and algorithms have made the Monte Carlo
ray-tracing methods an excellent choice for
most of the problems (Arvo et al., 2003).
In order to obtain glare data for any
given project, a new tool has been developed
to advance the current
state of the art
for solar reflectivity
studies using computational
fluid dynamics
(CFD). This proprietary
tool generates
glare data as follows:
• Using a custom
Carlo ray-tracing
the reflection
zone (see Figure
8) is estimated.
— Sun rays
are “injected” into the model at a
variable altitude and angle,
depending on the day of the year
and time of the day. A total of
500,000 rays are injected per
— The algorithm traces the rays’
trajectory and identifies the surfaces
that are impacted by incident
— Following the laws of optics, the
now reflected rays’ trajectory is
traced in the domain.
• Using a cell-face searching method,
the entire domain is explored, and
using geometry relationship only, we
can determine the path of reflected
• Glare intensity is calculated depending
on the distance, the direction,
and the concentration of the reflected
rays through a basic operation
that correlates the reflected light
zone from the ray calculation with
the glare intensity at each cell.
• Glare intensity is reported using the
custom fractional scale previously
described. Figure 9 shows an example
of the custom fractional scale on
the left-hand side.
The algorithm described above
provides glare data at a particular
time of day. This means that the
results are accurate for one particular
month, at one particular day, at a
given time of day. Therefore, several
iterations are required in order to create
representative data for the whole
calendar year.
The model domain will require
some level of accuracy with regard to
the surrounding environment. Adjacent
buildings may shade the project at various
periods of daylight on certain days of the
year. And conversely, roadways or neighboring
façades may be sensitive to reflections
bounced from the project’s reflective surfaces.
Engineering judgment will be needed
to quantify the scope of the domain model
and level of detail required to reasonably
predict areas of concern.
Models can be simplified in different
ways in order to accelerate computational
processes—especially since the algorithm
needs to be executed several times for a single
project. It is important to take into consideration
that this tool only captures primary
specular reflection.
This new technology can also be utilized
in determining glare conditions present in
airspace matters and not only ground level
glare. This type of application is used in
quantifying potential glare problems in airports’
runways, taxiways, control towers,
approach vectors, glide slopes, and with air
traffic. Figure 10 shows an example of glare
in airspace. Determining this type of glare is
helpful for new construction in or close to
airports. Vertical markers were added in
order to quantify the distance traveled by
the reflected light.
In order to analyze a project for potential
solar reflectivity issues, the following
information is required:
• Project location (latitude and longitude)
• 3-D “watertight” model of building,
including surrounding buildings
• Building’s orientation with respect
to true north
• Reflection coefficient and exterior
light for all exterior materials
Figure 7 – Glare examples.
Figure 8 – Sunlight injection representation.
S Y M P O S I U M O N B U I L D I N G E N V E L O P E T E C H N O L O G Y • OC T O B E R 2 0 1 2 C L I F T A N D MO N T E S – A M O R O S • 7 3
Reflected light can be looked at as a
source of light that originates at the building’s
glazed façade. As such, there is heat
associated with it that will be manifested
around the building’s domain. The amount
of heat that reflected rays originate will
depend on the number of reflected rays
coinciding within a particular area at a
given time. As a secondary step and using
the data generated with the solar reflectivity
analysis tool herewith presented, heat
differential (temperature increase) can also
be obtained.
When designing a building, it is very
important to consider the movement of the
sun in interaction with the design in question.
Factors that need to be carefully
designed and taken into consideration in
order to avoid solar reflectivity issues are
the following:
• Highly reflective glass
• South-facing concave building
shapes (for projects in the northern
• Elliptical building shapes in the vertical
• Change in planes throughout building
These are just a few factors to consider,
but location and neighboring buildings also
affect the path of reflected light and its
interaction with the entire project.
If designing for a downtown area or a
densely developed site, it is important to
consider street width, building orientation,
building height, and cladding materials. In
these cases, avoiding the potential for solar
reflectivity issues could represent a bigger
challenge, but it is something that can definitely
be avoided or mitigated.
This new CFD tool is valuable, as orientation
of the design can be easily rotated to
search for optimum results with respect to
mitigation of reflectivity. And the CFD’s provision
of a scale-of-intensity level is critical
to judge the limit of primary surface reflectivity
in regard to hazardous glare. With this
new technology, one can determine whether
or not the reflected glare is going to be a
In addition to glare intensity information,
the following can also be provided:
• Path of solar reflections
• Shadowing
• Solar data
• Temperature increase due to reflections
This can be applied to isolated buildings
or target buildings, including their surroundings.
Since the creation of this new tool and
its increasing demand, designers have
become aware of the importance of this type
of study. This tool has uncovered the need
for an industry-wide accepted criterion for
exterior glare, but it is too early to know
when this will be incorporated into design
codes, etc.
As architectural designs become
increasingly complex in shape and geometry,
the need for reflectivity studies is
heightened. This new CFD tool is available
to assist designers in making sound decisions
and avoiding pitfalls of poor performance.
J. Arvo, P. Dutre, A. Keller, H.W. Jensen,
A. Owen, M. Pharr, and P. Shirley,
2003. Monte Carlo Ray Tracing.
Siggraph 2003, Course 44, July 29,
D.N.H. Hassall, 1991. Reflectivity:
Dealing With Rogue Solar Reflections.
Published by author.
S. Naai-Jung and H. Yen-Shih, 2000,
“The Computer-Aided Visualization
of Curtain Wall Reflection Glare,”
ASCE Conference Proceedings 279,
205. August 14, 2000, Stanford, CA,
ASCE, pp. 1574-1581.
S. Naai-Jung and H. Yen-Shih, 2001a.
“A Study of Reflection Glare in
Taipei,” Building Research and
Information, 29:1, pp. 30-39.
S. Naai-Jung and H. Yen-Shih, 2001b.
“Volumetric Study of Reflected Glare
From Glass Curtain Walls of Buildings,”
Journal of Architectural Engineering,
Volume 7, Issue 3, pp. 87-
A. Rofail, B. Dowdle, J. Perry, 2004.
“Reflectivity Impact on Occupants of
Neighbouring Properties,” ICBEST
2004. Sydney, Australia.
R. Serway, 1997, Fisica, Tomo II. 4ta ed.
Mexico: McGraw-Hill.
P. Tippens, 1999. Fisica: Conceptos y
Aplicación, 5ta ed. Mexico: McGraw-
U.S. Department of Transportation –
Federal Aviation Administration,
2011. “Procedures for Handling Airspace
Matters,” Order JO 7400.2H.
Figure 9 – Glare analysis output example. Figure 10 – Glare in airspace, for airports.