Building and Environment 103 (2016) 193e202

Contents lists available at ScienceDirect

Building and Environment

journal homepage: www.elsevier.com/locate/buildenv

Assessing reected sunlight from building facades: A literature review

and proposed criteria
Ryan Danks a, *, Joel Good b, Ray Sinclair a
a
b

Rowan, Williams, Davies and Irwin (RWDI) Inc., 600 Southgate Dr., Guelph, ON, N1G 3W6, Canada
Rowan, Williams, Davies and Irwin (RWDI) Inc., 230 e 1385 West 8th Ave, Vancouver, BC, V5Z 1K5, Canada

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 23 February 2016
Received in revised form
28 March 2016
Accepted 12 April 2016
Available online 20 April 2016

In recent years there has been a growing awareness of the risks posed to people and property from
uncontrolled solar reections from the built environment. Despite the severity of the risks, there is
surprisingly little regulation regarding such reections. Presumably, part of the reason for this lack of
regulation is that there are no universally accepted criteria from the scientic community dening
acceptable limits of reected visible light and thermal irradiance in the urban realm. Without appropriate guidance, the regulations which are employed by cities may not be appropriate and designers have
no means to judge the impact of a potential building's reections until after its built. This paper presents
a review of existing regulations and metrics related to the impact of visible light and thermal energy on
people and property. It also proposes quantitative criteria which the authors have developed for use in
design and construction in order to help architects and designers understand the level of impact their
building's reections will have on its neighbors. The literature that the proposed criteria are based on is
still limited in breadth. It is our hope that the research and design communities will further develop the
criteria and tools that will benet designers and city regulators.
2016 Elsevier Ltd. All rights reserved.

Keywords:
Glare
Solar reection
Built environment
Regulation
Faade performance

1. Introduction
For people living in cities, seeing reected sunlight from a
building is a common experience, and while this can sometimes be
an annoyance, we mostly accept it as part of urban life. However,
buildings can cause more severe impacts owing to the type and
coverage of glass on the faade and also curvilinear designs. A wellpublicized example of signicant reections was the Disney Concert Hall in Los Angeles [1]. It had a signicant area of polished
stainless steel, which combined with the concave facade geometry,
would reect and sometimes focus light within the neighborhood.
This led to heat gain issues at sidewalks and inside surrounding
buildings. The bright reections also posed a visual distraction to
drivers. Ultimately, parts of the faade were roughened to help
scatter the reections, reducing their intensity. More recent examples of reected light causing distraction or damage include the

* Corresponding author.
E-mail addresses: ryan.danks@rwdi.com (R. Danks), joel.good@rwdi.com
(J. Good), ray.sinclair@rwdi.com (R. Sinclair).
http://dx.doi.org/10.1016/j.buildenv.2016.04.017
0360-1323/ 2016 Elsevier Ltd. All rights reserved.

Shard in London [2], Museum Tower in Dallas [3], the Vdara Hotel in
Las Vegas [4] and 20 Fenchurch in London [5]. The latter two
buildings reportedly focused enough sunlight in pedestrian areas to
cause burns on a guest and damage a car with radiant heat,
respectively. Buildings are not the only potential sources of
dangerous glare from the built environment; photovoltaic (PV)
panels [6] and even art installations [7] have been blamed for visual
and thermal issues related to reected sunlight as well.
The well-publicized nature of these events has led to increased
awareness in the design community, and a desire to understand
how a proposed building will interact with the sun. As consulting
engineers in the eld of building science, the authors have been
asked with increasing frequency: How will reections from my
building affect its neighbors? With little guidance available from
governing bodies we endeavored to create suitable criteria to assess
the visual and thermal impacts of reected light from the built
environment. These criteria were developed and rened through a
review of available literature across multiple scientic and engineering disciplines as well as through experience gained from the
analytical and physical study of reections from several buildings
around the world.

194

R. Danks et al. / Building and Environment 103 (2016) 193e202

2. Review of existing regulations regarding reected light

Despite the risks that glare can pose to people and property,
there is little regulation around it. The Federal Aviation Administration (FAA) in the United States has recently introduced provisions to reduce the risk of visual glare distracting pilots and air
trafc control (ATC) personnel [8]. However this is limited to potential glare from solar energy systems on airport property only.
German federal law classies light as an emission to the environment much like noise or exhaust [9] and a recent decision by the
German Federal/State Working Group for Pollution Control (LAI)
indicates that this includes reections [10]. It also denes any PV
system which causes glare for more than 30 minutes per day or
30 hours per year as a nuisance, but again this regulation does not
extend to facades in general. Many cities have guidelines related to
glare from building facades, but the majority of them only indicate
that an analysis of the glare impact should be undertaken, usually
at the discretion of the city. Often no clear guidelines on the surfaces to be studied, the criteria to be used or what constitutes
acceptability are dened. Only a few cities have implemented
prescriptive measures which aim to limit glare; typically by
restricting the use of faade materials which reect more than a
prescribed percentage of incident light. Several Australian cities
limit the normal reectance of any glass faade element to no more
than 20%. Sydney takes this one step further and applies the limit to
all faade materials [11]. Building authorities in Singapore, Hong
Kong and Shanghai set similar limits on faade reectivity [12e14]
but this type of regulation remains rare outside of southeast Asia
and Oceania. The city of Dallas, Texas is the only North American
city known to the authors that attempted to pass similar legislation,
but it was ultimately abandoned during the public consultation
phase. Prescriptive limits on faade material reectivity are
attractive regulations because they are easy to understand and
straightforward to enforce, however they overlook several important aspects of the physics of reected sunlight.
Firstly, using a single prescribed reectance limit fails to
acknowledge that the reectance of a surface is not a xed value.
When light moves from one medium to another, two things happen
at the interface; some light transmits through the interface into the
second medium and the rest is reected back in a new direction.
The fraction of light which is reected or transmitted can be
determined using the Fresnel Equations, which state that regardless of the indices of refraction of the two materials, the fraction of
reected light approaches 100% as the rays of light impact the
interface at more glancing angles. As an example, a single pane of
6 mm clear glass will reect approximately 8% of visible light for
light rays that strike perpendicular to the exterior surface of the
glass. For rays that strike within 50
of perpendicular, the reectance remains roughly constant; however, as the rays strike at angles approaching parallel to the reecting surface, the reectance
rises rapidly to 100%. This is illustrated graphically in Fig. 1 using
data generated by Lawrence Berkeley National Laboratory's WINDOW v7.2 software [15]. (In the chart, 0
indicates an incidence
angle perpendicular to the reecting surface.)
Glazing manufacturers typically only specify the reectance of
glazing for light striking it perpendicularly. This is frequently
misunderstood by designers who think that the low reectance
glazing will prevent any issues of glare. However, depending on the
orientation of the faade and the location of the sun in the sky, the
faade could be reecting signicantly more light than would be
expected based on the manufacturer's specication data.
Furthermore, these types of regulations often only consider how
reective the surface is to visible light and ignore how much of the
sun's thermal energy is reected. Many years ago this distinction
would not be signicant, as traditional windows reect visible and

thermal energy to approximately the same degree. Today, highperformance glazing systems are available which often employ
low-emissivity coatings. These products can be signicantly more
reective to thermal energy than they are for visible light. As an
example, consider the common Solarban 70XL glazing system,
manufactured by PPG Industries. According to the manufacturer
[16], typical double-pane glazing systems using Solarban 70XL,
reect between 5% and 21% of exterior visible light, which mostly
fall within the prescribed reectance limits discussed earlier. But,
when the glazing's reectance is averaged over the full solar
spectrum (i.e. ultraviolet, visible and infrared radiation), the
reectance of some systems can increase dramatically. In the case
of Solarban 70XL employed with two panes of clear glass, the full
spectrum reectance is more than four times the visible reectance. This is an important nuance as the heating of an object will
be driven by both visible and thermal energy. The effect that
Solarban 70XL has on the visible and full spectrum reectance of a
single pane of 6 mm glass is also illustrated in Fig. 1.
Another shortfall of prescriptive regulations is that they typically do not consider the nature of the impacts (i.e., the who or
what being impacted), nor the frequency and duration of the
reections. A visible reection that falls on a pedestrian could be
benign in many situations but when it aligns with a vehicle driver's
required line of sight and impairs their vision, it becomes a danger.
Similarly, a short duration reection is unlikely to cause signicant
heating or damage to surrounding vegetation or man-made objects,
but the longer the exposure, the higher risk of the potential adverse
thermal impacts.
Finally, to stipulate a general upper limit of reectance to any
one unit of a glazing assembly does not account for the building's
shape. In particular, a concave shaped facade can potentially
concentrate reections in one area and a convex shape can cause
repeated reections over a longer period of the day over a broader
area of the surrounding neighborhood. As faade designs become
more complex, these impacts become harder to intuitively predict.
For example, a single surface adhering to a 20% maximum reectance limit may not cause problems, but if a number of these individual surfaces create a focal point (e.g. a faade with a concave
curvature, illustrated in Fig. 2 above) the thermal effects would be
additive and could ultimately cause thermal discomfort or burns on
people or damage to plants and materials. Situations such as these
are often the culprit in cases of extreme reection related damage,
which have been dubbed death rays by the media.
Due to these limitations with existing prescriptive limits, it is
the authors' opinion that a more qualitative, analytical approach be
taken to assessing the impact of urban building reections.
3. Review of existing reection impact metrics
A current limitation with regulating urban reections is that
there is a lack of widely accepted metrics for dening the full
impact of a building's reections on its surroundings. This is at least
partially a result of the subjectivity of how an individual experiences a bright light. This experience is dependent on a variety of
factors including what the person is doing, their expectations, their
age, etc. Therefore, there is currently no universally accepted metric
for the impact of day light even within the transportation and
lighting communities.
A separate issue from visual impacts is the impact of reected
thermal energy. There is limited research available on how much
reected energy is too much from a building. Beyond the issue of
intense focused reected thermal energy and its impact on human
safety and material damage, are more subtle questions about human comfort and even additional cooling loads for adjacent
buildings.

R. Danks et al. / Building and Environment 103 (2016) 193e202

195

Fig. 1. Surface reectance as a function of light ray incidence angle.

Fig. 2. Schematic of a curved faade in elevation (left) and plan (right) focusing and concentrating reected energy.

3.1. Visual glare

Visual glare can broadly be divided into two types: disability and
discomfort glare. Disability glare occurs when an object in the eldof-view is bright enough to impair the ability of a viewer to
distinguish other objects due to a loss of contrast. This type of glare
is most often associated with road lighting, as the very large difference in brightness between road or automobile lighting and
ambient night-time light levels can very easily overwhelm the eye.
Discomfort glare is more often associated with interior lighting,
where a bright source of light exists in the eld-of-view, causing a
nuisance or annoyance for an individual.

3.1.1. Disability glare metrics

Several challenges exist in applying disability glare metrics
directly to exterior daytime glare situations. Firstly, within a city,
glare complaints tend to relate to discomfort glare rather than
disability [17]. Studies involving the impact of glare on drivers

[18,19] have also shown that glare sources which would not be
considered sources of disability glare can still degrade driver performance. Additionally, many metrics related to disability glare are
intended for night lighting of roads, and therefore are not necessarily applicable in a day time setting. As an example, the
commonly used Threshold Increment (TI) metric, which can be
thought of as the increase in an object's contrast required to
maintain visibility after a glare source is introduced, is only valid for
average road luminance levels between 0.05 and 5.0 cd/m2 [20].
During the day, typical road luminance values are several orders of
magnitude above this range [21], making this metric invalid for use
with solar reections. Despite these challenges, several metrics
have been proposed based on disability glare research.
One of the more common metrics in industry comes from
Hassall [22], where the so called veiling luminance (a measure of
disability glare) is limited to no more than 500 cd/m2 for motorists.
There is an assumption in this limit that glare only occurs during
late dawn or early twilight time frames, when a driver's eyes would

196

R. Danks et al. / Building and Environment 103 (2016) 193e202

be adapted to a 500 cd/m2 luminance level [23]. This assumption is

tied to his approach for determining the reected intensity. This
method involves the manual creation of sun-path diagrams for any
faade which may reect light, the areas where this light may
impact its surroundings is then determined, and the veiling luminance intensity is calculated by the Holladay formula [24]. Aside
from being cumbersome, particularly for complex facades and
multiple impact locations, this methodology is unable to dene the
length of time any given point experiences glare, and cannot easily
acknowledge the impact of multiple converging reections [25].
More recently, attempts have been made to update this methodology to employ modern computer software [26]. While this
eliminates much of the manual labor required, it does not address
many of the challenges noted above.
Jakubiec and Reinhart have proposed a limit of 30,000 cd/m2 for
exterior glare [27]. This is based on an assumed task luminance of
300 cd/m2, and the fact that the human eye can largely adapt to
between two and three orders of magnitude of luminance variation
in a given eld-of-view. However, this limit was derived based on
the assumed task luminance of viewing a computer monitor indoors. The wide variation in possible task luminance levels found
outdoors and the fact that many individuals have the ability to selfmitigate by looking away, limits this threshold's use in the general
case.
3.1.2. Discomfort glare metrics
Much like disability glare, discomfort glare also can be described
using many metrics. Unfortunately, there is an even greater lack of
consensus on which metric is most acceptable, particularly when
it comes to daylight applications. Part of the problem is that
discomfort glare is much more subjective than disability glare [28].
Therefore, the primary metrics that have been developed have been
based on correlations to survey results rather than physical measurements. These survey results were taken under indoor conditions and not under bright, outdoor sky conditions [29]. It is well
known that an individual's thermal discomfort is highly dependent
on their expectations, often more so than the actual environmental
conditions [30]. Stafford-Sewall et al. found that a similar issue
arises with night time glare impacts from headlights [31]. Their
work showed that complaints about being blinded by car headlights may not correlate well with their actual loss in visual acuity.
This effect is likely also present when it comes to visual discomfort
from daytime glare, and when coupled with the physiological adaptations the human eye can make to adapt to bright environments, it is unlikely that the results of indoor surveys will apply in
bright outdoor situations. This indicates a gap in the current body of
knowledge on this subject.
Investigations have shown that even under identical indoor
conditions the various glare metrics are inconsistent as they can
yield signicantly different results under day-lit conditions [32].
Several of the more common glare metrics, including the CIE Glare
Index (CGI) and the IESNA's Visual Comfort Probability (VCP) are
specically designed for interior situations where only small articial light sources are found, making them entirely unsuitable for
outdoor situations.
Another issue is that many of the discomfort metrics are based
on a contrast ratio (i.e. the ratio of the luminance of the glare source
to the overall luminance of the eld-of-view). Under bright daytime conditions, the luminance of the sun (or reections of the
sun) will dominate other sources of light, (as exemplied in Fig. 3),
forcing the ratio to close to 1. This scenario would result in the
prediction of a net reduction of discomfort for a viewer directly
looking at the glare, compared to a situation where the glare is
coming from further away from the line of sight. This is clearly not
correct.

Fig. 3. Example of reection in a bright outdoor context.

One metric, the Daylight Glare Probability (DGP) [33] does aim
to alleviate this problem by including the total illuminance on the
eye in addition to the luminance ratio described above in its glare
impact prediction. This improves the ability of the metric to
correctly identify extremely bright lights (i.e. the sun) as glare
sources. However, like the other discomfort glare metrics, DGP has
been correlated to survey data for individuals adapted to indoor
ofce-like conditions. Studies would be required in order to assess
whether it would be appropriate for individuals located in outdoor
environments, with signicantly brighter ambient lighting conditions and for people conducting signicantly different activities.
Particularly as some research has shown that even an individual's
thermal comfort can inuence their glare perception [34].
3.1.3. Other visual glare metrics
Studies into the physiological response of individuals to glare
have shown that the reexive physical changes of the human eye
when exposed to bright light correlate well with perceived
discomfort [35,36]. This type of approach is advantageous because
there is lower variability in the geometry and behavior of healthy
adult eyes under daytime conditions [37e39] compared to the
variability in the adaptation luminance (i.e. the brightness level the
eye is adapted to). Adaptation luminance denes what is considered impactful glare in most other metrics and depends on task and
the preferences of an individual in addition to physiology, making it
challenging to model generally [40]. While one would still have to
correlate the physical response to the external stimuli, this area of
research has promise for application to exterior reections as it has
the potential to at least partially reduce the subjectivity of visual
glare impact classication by taking individual preference out of
consideration.
Ho et al. [41] have formulated a visual glare risk metric based on
the potential of a glare source to cause after images in an observer's eld-of-view or, in more extreme cases, retinal damage.
Aside from the size and intensity of the glare source, the primary

R. Danks et al. / Building and Environment 103 (2016) 193e202

factors which dene this metric are the size of the pupil and the eye
as well as the speed of the human blink reex. As described above,
all these factors are relatively consistent for healthy, daylight
adapted adults. Most people have experienced the after-images
described by Ho when a camera ash is used in a dark room, which
also makes this metric very relatable for a lay-person compared to
some of the more subjective glare ranking scales [28]. This metric is
also easy to compute, allowing for rapid identication of problems
and mitigation tests. The use of this methodology is also explicitly
required by the Federal Aviation Administration as part of the
approval process for PV installations near airports [8] The main
limitations of the web-based tool also created by Ho [42] with
respect to its more general application, are that it does not identify
risks of multiple focused reections, or the specic location(s) of
the glare source(s). Also, the simplied geometrical analysis
employed in the tool ignores the impact of shading from the built
environment; which while appropriate for the open unobstructed
areas around airports, would be overly simplistic in an urban
context.
Attempts have also been made to leverage image processing
techniques to judge glare impacts on a viewer in terms of the
signal-to-noise ratio in a eld of view [43]. While this does have the
benet of being less subjective, further research is needed to dene
appropriate acceptability criteria.
3.2. Thermal impact of solar reections
In contrast to the plethora of metrics which assess the impacts
of visible light on individuals, there seems to be little if any publication of discussion of criteria for thermal energy reected from
buildings. To the authors' knowledge, the only available literature
regarding thermal irradiance limits is found in re safety or
military-related publications, referring to the impact on people and
property from the radiated energy of res or explosions.
3.2.1. Impacts on people
Fire safety research has provided a good understanding of how
long bare skin can be exposed to various levels of thermal radiation
before the onset of pain and then second degree burns. However,
there are many factors that can affect the severity and consequences of an exposure to elevated levels of thermal radiation.
Certainly, other environmental factors such as ambient temperature, clothing, age and health of person, and their ability to relocate
to less extreme exposure are just a few of the factors. Raj [44]
willingly exposed himself to 3500 W/m2 to 5000 W/m2 radiation
levels from a liqueed natural gas (LNG) re. He noted that with
exposures of several tens of seconds he did not suffer any severe
pain, burns or any other injury. He notes that American regulations
at that time had proposed an exposure limit of 1500 W/m2, for
which he argues are not justied due to attenuating effects of
typical clothing. The National Fire Protection Agency (NFPA) [45],
Society of Fire Protection Engineers (SFPE) [46] and British Standards Institution (BSI) [47] dene 2500 W/m2 as an upper limit of
acceptability for thermal radiation exposure during egress from a
re. They indicate that this level of intensity can be tolerated for
approximately 30 s before the onset of pain, allowing enough time
for an individual to evacuate. The Federal Emergency Management
Agency (FEMA) publishes a similar guideline for thermal irradiance
exposure [48].
The typical maximum level of solar irradiance individuals would
be exposed to naturally is approximately 1000 W/m2, and in the
authors' experience a typical intensity of a building's reection is in
the 200e400 W/m2 range, with peak intensities approaching
800 W/m2 possible. In contrast, Fig. 4 was produced during a study
of a building found to be focusing signicant amounts of solar

197

energy. It illustrates two focused glare events where the authors

directly measured irradiance levels both in and out of the focal area
to validate software predictions [49]. In this case we found that the
concave faade was routinely exposing areas to more than 3000 W/
m2. One member of the study team did voluntarily expose himself
to the focused reections while wearing long pants, a long-sleeved
shirt and a light jacket. He reported immediate thermal discomfort
and began sweating within seconds of exposure, despite the cool
ambient conditions. He removed himself from the reection in less
than 20 s due to the discomfort.
Thus, while re safety regulations provide a foundation of the
limits of exposure of people to thermal radiation from re, only in
the most extreme cases of focused reections do we see reection
intensities at or above these thresholds. Most importantly, however, is the question of what is an acceptable irradiance exposure
in an urban environment. It is not unreasonable to assume that
what one would consider acceptable while escaping from a re is
vastly different than what would be considered acceptable while
traversing an urban environment with no expectation of danger.
Therefore, the authors' opinion is that the re safety metrics can be
used as guidelines for upper limits but cannot be used directly, and
that a lower level of irradiance is likely appropriate for assessment
of reected sunlight issues.
There is also the risk of heat stress on individuals caused by
prolonged exposure to elevated temperatures, radiant ux and
humidity. The Wet Bulb Globe Temperature (WBGT) [50] is a
common metric used to provide a measure of an individual's
thermal stress from environmental conditions.
It is commonly used in sport, workplace safety and military
applications to determine the risk to athletes, military personnel, or
other workers performing strenuous activities in hot climates.
While there are well accepted limits based on WBGT, the authors
would again argue that they are not directly applicable in this
context for two reasons. The rst is that WBGT assumes a steadystate condition (i.e. that the radiative ux is constant), and therefore cannot easily acknowledge highly transient reections or the
fact that an individual would likely be able to move out of the way.
Secondly, is again the issue of what is an acceptable increase of
total radiant ux due to reected solar energy in terms of its effect
on the heat stress of an individual. The WBGT threshold stress
levels for workers, athletes or soldiers are not expected to be
applicable to an average pedestrian in an urban environment
exposed to reected sunlight.
Separate from the issue of direct exposure to thermal radiation
is that there can be concerns about people coming into contact with
objects which have been heated to relatively high temperatures by
reections. This is of particular concern if an individual does not
expect the surface to be hot because ambient conditions are cooler.
The National Aeronautics and Space Administration (NASA) denes
a surface temperature of 44
C (111
F) as the upper limit for a
surface which can be handled with bare hands for long periods
without the onset of pain [51].
Unfortunately, the temperature a surface reaches when exposed
to a reection depends on far more than just the intensity of that
reection. Material properties and ambient conditions also play a
large role in this complex and transient phenomenon.
As an example, Fig. 5 illustrates thermographic data of a concrete sidewalk and curb, as well as an asphalt drive surface exposed
to direct solar irradiance. Despite being exposed to the same level of
irradiance, the sidewalk, curb and asphalt all experience distinctly
different temperature gains. This difference is driven by the
different materiality between the asphalt and concrete as well as
the differing thickness (and thus thermal mass) of the curb
compared to the sidewalk. Predicting surface temperature gains is
made difcult by the highly complex wind ows, solar exposures

198

R. Danks et al. / Building and Environment 103 (2016) 193e202

Fig. 4. Plots of measured and simulated irradiance levels on a horizontal surface for case study with focusing reections.

Fig. 5. Visible light and thermographic photographs of surfaces exposed to direct solar irradiance.

and the varied materials found in a typical urban environment.

Therefore, temperature based restrictions are not suitable criteria
for design, in general. Thus, the authors recommend that criteria for
direct exposure to irradiance be the guiding factor in determining a
thermal tolerance limit.

3.2.2. Impacts on property and landscaping

Similar to the impacts of thermal irradiance on people, research
into the impacts on property is also mostly conned to the context
of re safety. The issue investigated by such studies is typically the
irradiance required for charring, ignition or combustion of various
materials.
Generally, the level of irradiance required to ignite common
materials in the presence of a pilot ame is at least 8000 W/m2 for
10 min [52], with auto-ignition possible at irradiances of
16,000e25,000 W/m2. These levels of irradiance would be rare in
the context of reected sunlight and far exceed the limits imposed
by human safety. That being said, many materials can soften and/or
deform at much lower irradiance levels if its temperature rises
above a critical value and/or is under an applied force (including its
own weight). As an example, there have been multiple recorded
instances where solar reections from residential windows were
able to heat vinyl siding to the point of warping [53]. Such critical
temperatures are well known by manufacturers; however as noted

above, determining a priori temperature gains due to reections in

a complex urban environment can have signicant uncertainty and
therefore is not an appropriate metric for design. Additionally,
there also appears to be a gap in the published knowledge base
regarding the impacts of lower thermal irradiance exposure on
common materials in real-world settings.
An additional concern of reected solar energy is the potential
negative impact on plant life. Generally speaking, plants are able to
protect themselves during times of excessive sunlight exposure
during the growing season, so long as they can sufciently regulate
their temperature via evaporation from leaves or convection to the
air to prevent physical damage [54]. Studies have also shown that
altering daylight exposure patterns in laboratories can have a
physical impact on plants [55]. During the literature search we
could not nd scientic research of sunlight reections from the
built environment in outdoor settings where naturally occurring
sunlight will also be present at levels adversely affecting plants. The
authors' experience is that in an outdoor urban environment plants
are often exposed to peak solar irradiance levels more intense than
most building reections. Also urban plants experience high variability of environmental factors like shade, dry bulb temperature,
and wind speed. As plants have shown to be resilient next to other
reective surfaces (e.g. plants along a shoreline of a lake) it is
thought that only focused reection scenarios would cause

R. Danks et al. / Building and Environment 103 (2016) 193e202

problems. The challenge in assessing the adverse impact of reections on plants is that they can be become unhealthy from a
wide variety of other factors, or a combination of factors, including:
excessive/insufcient pruning or mowing, pedestrian trafc, over/
under watering, poor quality soil, poor ventilation, and infestation
of insects, molds, or fungi. Moreover, the degree of the adverse
effects would vary by species of plant. This gap in current knowledge limits our ability to establish limiting criteria on reection
impacts on plants.
3.3. Additional metrics
In addition to metrics which intend to quantify the specic
impacts of visual and thermal reections, there are also other
proposed metrics which try to provide an indication of the reection impacts of a building in a more general manner. These types of
metrics are typically geometric in nature, using only the geometric
relationship between the sun and the reecting building to dene
the impact of reections. Shih and Huang [56] dene a boundary
of reection area (BRA) parameter, which is essentially the area of
a horizontal grade level plane which is exposed to reections from
the subject building. While simple to understand and compute, this
method fails to acknowledge the impact of variable glazing
reectivity, shading and the risk of focused reections. Brzezicki
suggests a similar analytic approach to provide an early warning
of potential reection focusing from curved facades [57]. As with
the BRA, while easy to compute, this approach ignores the variable
reectivity of common faade elements, shading and reections
from planar surfaces.
Another approach is to use forward ray tracing methods (i.e. a
computer simulation which tracks the paths of light rays starting at
the sun and moving forward in time) to identify areas within a
study domain where reections land, and then quantify the
reection impact in terms of the density of rays which impact a
given area [58], or in terms of the reected energy expressed as a
multiple of the incident light [59]. While the density of reected
rays impacting a surface does identify locations where focused,
frequent or long duration reections occur, it gives no indication of
the overall intensity of those reections. Similarly, proposing limits
on reected light to no more than ambient (as in Ref. [59]) can be
misleading. Ambient solar intensity varies over the course of a day
and throughout the year, a reection which is twice as intense as
the sun at dawn may be an order of magnitude less intense than a
reection which is twice as intense as the sun at noon and thus
have vastly different visual and thermal consequences for people
and property exposed to it.
4. Proposed criteria
The following sections dene a set of criteria that the authors
propose are suitable for dening the level of impact a reection will
have on its surroundings. The use of the proposed visual light and
thermal radiation criteria presented here presuppose that a reliable
method of measuring and/or predicting reected irradiance at a
given point has been used as an input. A review of the various
potential pieces of software which can be used to calculate reected
irradiance is a topic worthy of its own discussion, which the authors
plan to address in an upcoming publication [49].
The impacts of visual reections will occur at much lower intensities than those which can cause thermal impacts; also many
materials (particularly high performance architectural glazing) will
reect different amounts of visible and thermal irradiance. Therefore, we believe it is appropriate to dene separate criteria for the
visual and thermal impacts of reected solar energy to properly
understand both types of effects.

199

4.1. Visible glare

Based on a review of existing metrics the authors could nd no
single metric which we believe fully captures how visible reected
light from a building would impact an individual. However, our
opinion is the metric put forth by Ho et al. [41] is currently the best
practical measure when combined with some other heuristics.
The Ho et al. metric is explicitly designed and tested for outdoor
use in the presence of solar reections. Unlike the other metrics, it
is also based on human physiology, which as noted above, is less
subjective than preference based metrics and the after-imaging it
is based on is also easy to explain to lay-people. It is straightforward
to compute and is not tied to a specic piece of software. An added
advantage is that the Ho methodology is already required by the
FAA as part of the approval process for PV panels on airport land, so
applying more broadly would ensure a consistent approach and
terminology for all outdoor glare across multiple domains.
Since this metric was originally designed to test for impacts on
people engaged in safety critical tasks (i.e. ying a plane or working
in an air trafc control tower) it makes several assumptions which
may be overly conservative when applied in a broader context.
Specically, the metric assumes any reection bright enough to
cause after imaging is problematic, which implies that the viewer is
always directly viewing the source of glare. To update this for an
urban reection context, we propose combining the original three
categories dened by Ho et al. with the location of the glare source
and the assumed activity of the viewer.
We dene two types of activities in this context; low risk activities are those where temporary visual impairment due to glare
within the eld-of-view does not pose a safety risk to the viewer or
others around them. Examples would be pedestrians walking midblock who can easily avert their gaze until they have moved out of
the reection, those sitting outdoors who can move their position
and orientation, and workers in adjacent buildings who can control
window shades. Conversely, high risk activities are those where a
xed view direction is required and any after-image causing reections within that eld-of-view pose a safety risk to the viewer
and those around them. Examples of what the authors would
consider high risk activities include driving cars, ying planes and
any other task where an individual losing visual acuity puts lives at
risk.
Once a task has been assumed, one can then determine the role
that the location of the glare source will have on its impact. Since
those engaged in low risk activities can, by denition easily selfmitigate the glare impact, the location of the reection is irrelevant for those locations.
Those engaged in high risk activity on the other hand, have a
required eld-of-view in order to safely engage in the activity. We
would categorize reections from outside of this required eld-ofview similarly as low risk activities, that is to say, a nuisance rather
than a safety risk.
The angular radius which denes the required eld-of-view will
vary depending on the assumed task. For drivers, a 20
radius is
recommended because for a typical driver looking straight ahead;
view angles beyond 20
tend to be obscured by the dashboard, roof,
and other opaque car body elements [60]. Tests on pilots have
shown that glare within a similar 25
radius had the greatest
impact on perceived performance, while some impact still occurred
for glare from as far away as 50
[61]. Some research has also been
conducted into glare impacts on train operators [62] and operators
of construction equipment [63], but further validation and investigation of these topics is likely required.
Using the above criteria we can then generalize the glare impact
categories dened by Ho et al. by adding a fourth category to
differentiate the impacts of after-image causing glare on those

200

R. Danks et al. / Building and Environment 103 (2016) 193e202

Table 1
Proposed visual impact categorization.
Visual impact
level

Reection conditions

None
Low
Moderate

No reections impact the point

Reections are either dim or small enough as to present a low risk of causing after-images for an observer
Reections are bright and/or large enough to potentially cause after-images in an observer, and the observer is engaged in a low risk activity (e.g.
walking), or the observer is engaged in a high risk activity (e.g. driving) and the source of the reection falls outside of the observer's required eld-ofview.
Reections are bright and/or large enough to potentially cause after-images in an observer, and the observer is engaged in a high risk activity (e.g.
driving) and the source of the reection falls within the observer's required eld-of-view.
Reections which are bright and/or large enough to potentially cause retinal burns.

High
Extreme

engaged in different types of activities and those not directly facing

the source of glare.
Table 1 summarizes our proposed categorization of visual glare
impacts.
These proposed criteria can be applied to any location around a
building. For a selected view point, the results can be graphically
displayed using annual plots such as the type used by the authors as
illustrated in Fig. 6. These graphics can easily convey reection
frequency, duration, source, and associated impact level. The
None category is purposely combined with the Low category in
such plots to better highlight only the potentially problematic
reections.
Engineering judgment is currently employed to communicate
the overall glare impact at a location, and if mitigation of the glare is
recommended based on the level of impact, timing, frequency, and
duration of individual glare events.
The authors' opinion through our work with designers is that, in
general, any point location that experiences high-impact events for
more than 5 minutes at a time, for more than 2 weeks out of the
year would be classied as having a high overall visual impact. In
this case, mitigation measures would be recommended for the
source of the glare. The overall visual impact at a point only affected
by moderate glare requires a more nuanced approach which must
be evaluated on a case-by-case basis. This is the most pragmatic
approach until further research better quanties appropriate values
for the above parameters. However, since many jurisdictions do not
have strong, prescriptive glare requirements; the nal decision to
implement any mitigation ultimately depends on the level of risk
that the building owner/developer/designers are willing to accept.
4.2. Thermal impact of solar reections
As noted in section 3.2, the thermal impacts of reected solar
energy are not currently well dened. The available literature is
predominantly related to re safety. The authors are not aware of
any studies which have determined the difference of impacts of
exposure from re and sunlight. Therefore, the criteria we apply for
impacts of focused sunlight are based on published criteria from
the re safety literature. It is unlikely, given the geometry of the
human body, that an individual would be exposed to the full intensity of direct irradiance from the sun and a focused reection at
the same location on their body. Thus, we apply the criteria to the
reected irradiance only.
According to NFPA 130 [45] and the SPFE [46], thermal irradiance levels above 2500 W/m2 can lead to the onset of pain and skin

damage within 30 s of exposure. This level of heat ux can also raise

the temperature of surfaces which people may come into contact
with, posing additional risk to human safety. Therefore, reections
with intensities higher than this should not be permitted in any
spaces where people may reside, which include not just grade level
spaces (e.g. parks, sidewalks, etc.), but also building facades (since
reections can readily transmit into spaces via windows), and roof
tops. Borrowing nomenclature from the eld of occupational health
and safety, we dene this level of irradiance as the ceiling exposure
limit, i.e. the maximum exposure one can be subjected to for any
length of time.
Lower reection intensities, while not posing as serious of a risk
to human safety, can still negatively impact human comfort. There
are no denitive guidelines or criteria with respect to this issue. We
know this criterion should be less than 2500 W/m2 and greater
than typical peak solar noon levels of 1000 W/m2 which people
commonly experience. Our opinion at this time is that a reasonable
criterion may be to limit reected irradiance exposure to 1500 W/
m2 or less. Based on our assessment, we believe at this level of
irradiance most people would be able to tolerate it for several minutes before the onset of discomfort. As an example, a 1500 W/m2
reection impacting person under typical ambient conditions (i.e.
25
C, 50% relative humidity, a 1.5 m/s wind speed, and 1000 W/m2
of solar insolation) would lead to a 30% increase in WBGT (33
C
compared to 25
C). While this value does exceed typical WBGT
thresholds [64], it is important to remember the transient nature of
reections and the steady-state nature of the WBGT calculation.
The Occupational Safety & Health Administration (OSHA) indicates
that time-weighted averages of WBGT values over a 120 minute
period should be used for intermittent exposures [64]. Even if the
individual was exposed to the 1500 W/m2 reection for 60 min, the
time-weighted WBGT value would be 29
C, which remains well
below the OSHA recommended threshold. Additionally reections
at this intensity level will heat surfaces more slowly. Thus we feel
reections below 1500 W/m2 pose a reduced risk to people and
should therefore be considered a short term exposure limit. We
would conservatively dene short term as 10 minutes or less
which is slightly shorter than the standard 15 minute denition of
short term used in the occupational safety context.
With respect to the impacts on plant life, there is currently
insufcient data on the impacts of intermittent solar reections on
plant life in a realistic setting. This makes it difcult to determine an
appropriate threshold to limit irradiance to prevent damage to
plants. From a practical perspective in interpreting the results of
the predicted impacts of reected light we would normally identify

Table 2
Proposed thermal limits.
Condition

Suggested thermal irradiance threshold

Short term exposure limit (authors' opinion)

Ceiling exposure limit (NFPA, SFPE literature)

1500 W/m2
2500 W/m2

R. Danks et al. / Building and Environment 103 (2016) 193e202

201

Fig. 6. Sample annual glare impact plot.

potential concerns about long duration impacts of reected light at

levels greater than 1000 W/m2 and more signicant concerns about
impacts on plants for short-term impacts of irradiance levels above
1500 W/m2.
The proposed thermal limits can be graphically presented
in a similar fashion as the visual criteria shown in Fig. 6, clearly
conveying annual exceedances for any desired point location. Table
2 summarizes our proposed limits for exposure to reected thermal
irradiance.
5. Conclusions
The authors have summarized a review of the current state-ofthe-art in measuring and regulating the impacts of reected solar
energy in the built environment. Despite the potential risk associated with uncontrolled reections, there is little regulation surrounding them. Part of the reason for this is that there is no
universally accepted set of criteria dening acceptable limits of
visible and thermal impacts in the urban realm from the scientic
community.
With the current lack of empirical research on this topic, the
authors have put forward what we believe to be pragmatic criteria,
suitable for design purposes and post-construction assessments.
Some of the criteria proposed are based on interpretation of sparse
research as well as the authors' opinions and experience and, as
such, are open to feedback. This work will hopefully stimulate
further research and publications that will lead to better substantiated design criteria. Until then, the criteria presented herein are
useful for real world design.
6. Suggestions for future work
As noted throughout this paper, there is a signicant lack of
direct research on the impact of visible and thermal reections
from the built environment on people and property. Additional
areas of research that have been highlighted to help close this gap
include:
 Improved quantication of the impact of outdoor daytime glare
from solar reections on the ability of drivers, pilots and other
operators of heavy machinery to control their vehicles in terms
of reection intensity, percentage of visual eld, and duration.
 Quantication of the acceptability of reections from a human
vision and thermal comfort standpoint for individuals in an
urban environment.
 Quantication of the upper thermal limits on people in more
typical contexts (i.e. non re scenarios). This could include

determining acceptable exposure times for occupants not

expecting a sudden heat gain, and the thermal limits for short
and longer exposure times. Ideally, this would establish both
safety and comfort limits.
 Investigations into the frequency, intensity and durations
required for reected irradiance to cause signicant heat gain
issues within an adjacent buildings for various wall
constructions.
 Measurements of surface temperature gains of common objects
under various combinations of environmental conditions and
reection intensities.
 Quantication of the impacts reected light has on plant life
though heating or disruption of natural circadian rhythms.
Beyond the building science applications, this research also has
direct applications in the elds of horticulture, ergonomics, optics,
ophthalmology, as well as transportation and occupational safety.
Funding
This research received no specic grant from any funding
agency in the public, commercial, or not-for-prot sectors.
Acknowledgments
The authors thank those who provided insightful comments and
suggestions on this text. We would also like to acknowledge the
support of our employer during the development and publication
of this work.
References
[1] M. Schiler, Examples of glare remediation techniques, in: PLEA 2009: Proc.
26th Conf. Passiv. Low Energy Archit., Quebec City, 2009.
[2] A.M. City, Dazzling Shard Has Inadvertent Effect of Blinding Train Drivers, 19
September 2012 [Online]. Available: http://www.cityam.com/article/dazzlingshard-has-inadvertent-effect-blinding-train-drivers [Accessed 27 March
2015].
[3] M. Granberry, Museum Tower Hires Technical Firm to Help Solve Glare
Dispute e Dallas Morning News, 3 April 2014 [Online]. Available: http://www.
dallasnews.com/lifestyles/arts/columnists/michael-granberry/20140402museum-tower-hires-technical-rm-to-help-solve-glare-dispute.ece
[Accessed 27 March 2015].
[4] NBC News, 'Death Ray' at Vegas Hotel Pool Heats up Guests, 30 September
2010 [Online]. Available: http://www.nbcnews.com/id/39403349/ns/travelnews/t/death-ray-vegas-hotel-pool-heats-guests/#.VdcnWNVVhBc [Accessed
28 March 2015].
[5] O. Wainwright, Walkie Talkie architect 'didn't realize it was going to be so
hot', 6 September 2013 [Online]. Available: http://www.theguardian.com/
artanddesign/2013/sep/06/walkie-talkie-architect-predicted-reection-sunrays [Accessed 22 July 2015].
[6] M. Hayward, Airport Controllers Complain of Solar Panels' Glare, 30 August

202

[7]

[8]

[9]
[10]
[11]
[12]
[13]
[14]

[15]
[16]

[17]
[18]

[19]

[20]
[21]
[22]
[23]

[24]

[25]
[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

R. Danks et al. / Building and Environment 103 (2016) 193e202

2012 [Online]. Available: http://www.unionleader.com/article/20120830/
NEWS02/708309966/0/SPORTS.
J. Markusoff, When Art Attacks: City Pulls Sculpture after Visitor's Jacket
Singed by Reected Rays, 8 October 2014 [Online]. Available: http://www.
calgaryherald.com/news/calgary/whenattackscitypullssculptureafter
visitorjacketsingedreectedrays/10276470/story.html [Accessed 20
February 2015].
FAA, Interim Policy, FAA Review of Solar Energy Systems on Federally Obligated Airports, 78 FR 63276, United States Department of Transportation/
Federal Aviation Administration, 2013.
German government, Federal Immission Control Act, 2009.
LAI, Hinweise zur Messung, Beurteilung und Minderung von Lichtimmissionder-Arbeitsgemeinschaft fr Immissionsschutz, Berlin, 2012.
nen, Bund/La
City of Syndey, Central Sydney Development Control Plan, 2012. Sydney.
City of Singapore, Building Control Act, Chapter 29, Building Control Regulations, 2003. Singapore.
Hong Kong Building Department, Guidelines on Design and Construction
Requirements for Energy Efcient of Residential Buildings, 2014. Hong Kong.
Shanghai Urban and Rural Construction and Transportation Commission,
DGJ08-56-2012: Technical Code for Building Curtain Wall, Shanghai Construction and Transportation Committee, Shanghai, 2012.
LBNL, WINDOW Vers. 7.3, Lawrence Berkeley National Laborarory, 2015.
PPG Industries, Solarban 70XL Datasheet, 28 March 2015 [Online]. Available:
http://www.ppgideascapes.com/getmedia/4c819437-243f-48f1-8050eb88e22e4a70/10106Solaraban70XL7097.pdf.aspx [Accessed 20 April 2015].
N.-J. Shih, Y.-S. Huang, A study of reection glare in Taipei, Build. Res. Inf. 29
(1) (2001) 30e39, http://dx.doi.org/10.1080/09613210150208778.
R. Grey, D. Regan, Glare susceptibility test results correlate with temporal
safety margin when executing turns across approaching vehicles in simulated
low sun conditions, Ophthalmic Physiol. Opt. 27 (1) (2007) 440e450, http://
dx.doi.org/10.1111/j.1475-1313.2007.00503.x.
J. Theeuwes, J.W. Alferdinck, M. Perel, Relation between glare and driving
performance, Hum. Factors 44 (1) (2002) 95e107, http://dx.doi.org/10.1518/
0018720024494775.
CIE, CIE 140-2000: Road Lighting Calculations, Commission internationale de
clairage, Vienna, 2000.
l'e
BSI, BS 5489-2:2003: Code of Practice for the Design of Road Lighting, British
Standards Institute, London, 2003.
D. Hassall, Reectivity: Dealing with Rogue Solar Reections, Faculty of Architecture, University of New South Wales, Sydney, 1991.
F. Schieber, M. Harms, Analytic Study of Daytime Running Lights as Potential
Sources of Disability and Discomfort Glare under Ambient Illumination Conditions Ranging from Dawn through Dusk, Heimstra Human Factors Laboratories, University of South Dakota, Vermillion, 1998.
J.J. Vos, B.L. Cole, H.W. Bodmann, E. Colombo, T. Takeuchi, T.J. van den Berg,
CIE 146e2002: CIE Equations for Disability Glare, Commission internationale
clairage, Vienna, 2002.
de l'e
A. Rofail, B. Dowdle, J. Perry, Reectivity impact on occupants of neighbouring
properties, in: Proc. Int. Conf. Build. Envel. Syst. Technol., Syndey, 2004.
V.L.-R. Gil, Evaluation of solar glare from reective facades: a general method,
Light.
Res.
Technol.
(2015)
1e18,
http://dx.doi.org/10.1177/
1477153515574144.
J.A. Jakubiec, C. Reinhart, Assessing disability glare potential of reections
from new construction: case study analysis and recommendations for the
future, Transp. Res. Rec. 2449 (2014) 114e122, http://dx.doi.org/10.3141/
2449-13.
S. Fotios, Research note: uncertainty in subjective evaluation of discomfort
glare, Light. Res. Technol. 47 (2015) 379e383, http://dx.doi.org/10.1177/
1477153515574985.
R. Harrold, D. Mennie, IESNA Lighting Ready Reference: a Compendium of
Materials from the IESNA Lighting Handbook e Lighting Fundamentals, Illuminating Engineering Society North America, New York, 2003.
ppe, Different aspects of assessing indoor and outdoor thermal comfort,
P. Ho
Energy Build. 34 (6) (2002) 661e665, http://dx.doi.org/10.1016/S03787788(02)00017-8.
A.A. Stafford Sewall, S.A. Whetsel Borzendowski, R.A. Tyrrell, B.R. Stephens,
P.J. Rosopa, Observers' judgments of the effects of glare on their visual acuity
for high and low contrast stimuli, Perception (2016), http://dx.doi.org/
10.1177/0301006616633591 (in press).
J.A. Jakubiec, C. Reinhart, The adaptive zone e a concept for assessing
discomfort glare throughout daylit spaces, Light. Res. Technol. 44 (2) (2012)
149e170, http://dx.doi.org/10.1177/1477153511420097.
J. Weinold, J. Christoffersen, Evaluation methods and development of a new
glare prediction model for daylight environments with the use of CCD cameras and RADIANCE, Energy Build. 38 (7) (2006) 743e757, http://dx.doi.org/
10.1016/j.enbuild.2006.03.017.
n, R. Rodriguez, A. Pattini, Effects of perceived indoor temperature
J.Y. Garreto
on daylight glare perception, Build. Res. Inf. (2015) 1e14, http://dx.doi.org/
10.1080/09613218.2016.1103116.
Y. Lin, S. Fotios, M. Wei, Y. Liu, W. Guo, Y. Sun, Eye movement and pupil size
constriction under discomfort glare, Investig. Ophthalmol. Vis. Sci. 56 (3)
(2015) 1649e1656, http://dx.doi.org/10.1167/iovs.14-15963.

 n, R. Rodriguez, A. Ruiz, A. Pattini, Degree of eye opening: a new

[36] J.Y. Garreto
discomfort glare indicator, Build. Environ. 88 (2015) 142e150, http://
dx.doi.org/10.1016/j.buildenv.2014.11.010.
[37] A.B. Watson, J.I. Yellot, A unied formula for light-adapted pupil size, J. Vis. 12
(10) (2012) 1e16, http://dx.doi.org/10.1167/12.10.12.
[38] I. Bekerman, P. Gottlieb, M. Vaiman, Variations in eyeball diameters of healthy
adults, J. Ophthalmol. 2014 (2014), http://dx.doi.org/10.1155/2014/503645.
rzin, A.A. Barreira, Blink
[39] P.A.T. Kimaid, L.A.L. Resende, H.A. de Lima Castro, F. Be
reex: comparison of latency measurements in different human races, Arq.
Neuropsiquiatr. 60 (3) (2012) 563e565, http://dx.doi.org/10.1590/S0004282X2002000400009.
[40] J.A. Ferwerda, S.N. Pattanaik, P. Shriley, D.P. Greenberg, Model of visual
adaptation for realistic image synthesis, in: SIGGRAPH '96: Proc. 23rd Annu.
Conf. Comput. Graph. Interact. Tech., New Orleans, 1996.
[41] C. Ho, C. Ghanbari, R. Diver, Methodology to assess potential glint and glare
hazards from concentrating solar power plants: analytical models and
experimental validation, J. Sol. Energy Eng. 133 (3) (2011), http://dx.doi.org/
10.1115/1.4004349, 031021-1-031022-9.
[42] C. Ho, C. Sims, J. Yellowhair, Solar Glare Hazard Analysis Tool (SGHAT) User's
Manual v. 2H, Sandia National Laboratories, 2013.
[43] Y. Ou, Quantitative study of reection of sunlight by a glass curtain wall
resulting in a visual masking effect, Appl. Opt. 53 (29) (2014) 6893e6899,
http://dx.doi.org/10.1364/AO.53.006893.
[44] P.K. Raj, Field tests on human tolerance to (LNG) re radiant heat exposure,
and attenuation effects of clothing and other objects, J. Hazard. Mater. 157
(2e3) (2008) 247e259, http://dx.doi.org/10.1016/j.jhazmat.2007.12.114.
[45] NFPA, NFPA 130: Standard for Fixed Guideway Transit and Passenger Rail
Systems, National Fire Protection Association, Quincy, 2014.
[46] P.J. DiNenno, SFPE Handbook of Fire Protection Engineering, Society of Fire
Protection Engineers, 2002.
[47] BSI, PD 7974e6:2004: the Application of Fire Safety Engineering Priciples to
Fire Safety Design of Buildings - Part 6: Human Factors, British Standards
Institute, London, 2004.
[48] FEMA, Handbook of Chemical Hazard Analysis Procedures, Federal Emergency
Management Agency Publications Ofce, Washington D.C., 1988.
[49] R. Danks, J. Good, Urban scale simulations of solar reections in the built
environment: methodology & validation, in: 2016 Proc. Symp. Simul. Archit.
Urban Des., London, 2016.
[50] K. Parsons, Heat stress standard ISO 7243 and its global application, Ind.
Health 44 (3) (2006) 368e379, http://dx.doi.org/10.2486/indhealth.44.368.
[51] E. Ungar, K. Stroud, A new approach to dening human touch temperature
standards, in: Proc. 40th Int. Conf. Environ. Syst., Barcelona, 2010.
[52] BRE, PD 2552: Fire Spread in Car Parks, Building Research Establishment,
London, 2010.
[53] R. Hart, C. Curcija, D. Arasteh, H. Goudey, C. Kohler, S. Selkowitz, Research
Needs: Glass Solar Reectance and Vinyl Siding LBNL-5022E, Lawrence Berkeley National Laboratory, 2011.
[54] N.E. Holt, D. Zigmantas, L. Valkunas, X.-P. Li, K.K. Niyogi, G.R. Flemming,
Carotenoid cation formation and the regulation of photosynthetic light harvesting, Science 307 (5708) (2005) 433e436, http://dx.doi.org/10.1126/
science.1105833.
[55] M.J. Dowson-Day, A.J. Millar, Circadian dysfunction causes aberrant hypocotyl
elongation patterns in Arabidopsis, Plant J. 17 (1) (1999) 63e71, http://
dx.doi.org/10.1046/j.1365-313X.1999.00353.x.
[56] N.-J. Shih, Y.-S. Huang, An analysis and simulation of curtain wall reection
glare, Build. Environ. 36 (5) (2001) 619e626, http://dx.doi.org/10.1016/
S0360-1323(00)00034-2.
[57] M. Brzezicki, The inuence of reected solar glare caused by the glass cladding
of a building: application of caustic curve analysis, Comput. Aided Civ. Infrastruct. Eng. 27 (5) (2012) 347e357, http://dx.doi.org/10.1111/j.14678667.2011.00751.x.
[58] X. Yang, L. Grobe, S. Wittkopf, Simulation of reected daylight from building
envelopes, in: Proc. BS2013: 13th Conf. of Int. Build. Perform. Simul. Assoc.,
ry, 2013.
Chambe
[59] C.D. Clift, V. Montes-Amoros, Solar reectivity studies, in: Build. Envel.
Technol. Symp. 2012 Proc., Phoenix, 2012.
rez, Visual elds at the wheel, Optom. Vis. Sci.
[60] F. Vargas-Martin, M.A. Garca-Pe
82
(8)
(2005)
675e681,
http://dx.doi.org/10.1097/
01.opx.0000175624.34252.73.
[61] J.A. Rodgers, C.K. Ho, A. Mead, A. Millan, M. Beben, G. Drechsler, Evaluation of
Glare as a Hazard for General Aviation Pilots on Final Approach, Federal
Aviation Administration, DOT/FAA/AM-15/12, Ofce of Aerspace Medicine,
Washington D.C., 2015.
[62] D.P. Jenkins, L.M. Baker, C. Harvey, A practical approach to glare assessment
for train cabs, Appl. Ergon. 47 (2015) 170e180, http://dx.doi.org/10.1016/
j.apergo.2014.09.010.
[63] J.W. Hinze, J. Teizer, Visibility-related fatalities related to construction
equipment, Saf. Sci. 49 (2011) 709e718, http://dx.doi.org/10.1016/
j.ssci.2011.01.007.
[64] OSHA, OSHA Technical Manual (Section III: Chapter 4-Heat Stress), Occupational Safety and Health Administration, 1996.