University of Southern California Dissertations and Theses

Blind to light loss: evaluating light loss through commercial building facades as a contribution to urban light pollution

(USC Thesis Other)

Blind to light loss: evaluating light loss through commercial building facades as a contribution to urban light pollution

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Blind to Light Loss:
Evaluating Light Loss through Commercial Building Facades
as a Contribution to Urban Light Pollution

By
Dennis Joseph Chow

Presented to the
FACULTY OF THE
SCHOOL OF ARCHITECTURE
UNIVERSITY OF SOUTHERN CALIFORNIA
In partial fulfillment of the
Requirements of degree
MASTER OF BUILDING SCIENCE

AUGUST 2016

Chow | 2

COMMITTEE
Douglas Noble, FAIA, Ph.D.
Associate Professor
USC School of Architecture
dnoble@usc.edu
(213) 740-2723

Travis Longcore, Ph.D., GISP
Assistant Professor
USC School of Architecture
longcore@usc.edu
(310) 247-9719

Marc Schiler, LC, FASES
Professor
USC School of Architecture
marcs@usc.edu
(213) 740-4591

Chow | 3

ABSTRACT
In the discipline of building science, considerable effort is expended to evaluate the infiltration of light through a commercial
building facade. The designer/engineer must mitigate daylight through glazing, and must combat issues of glare and internal
heat gain. After dark, roadway and architectural lighting spill into the environment as urban centers become their own
sources of illumination. The effect is increasing night sky brightness, glare, light trespass, and other light pollution
phenomena at a global scale.
With increasing attention to the effects of light pollution on the fields of astronomy, ecology, medicine, and design,
Architects and Engineers may consider the alternative path of travel of light through a facade and imagine the degree of loss
reduction due to infiltration mitigation strategies. To answer these and other questions, a methodology was developed to
gather case study data from the 800 Wilshire Blvd. office tower in Downtown Los Angeles. To measure interior light loss
through the building facade, luminance and illuminance were evaluated using an HDR-capable camera outfitted with a fish-
eye lens and an illuminance meter. These measurements were used to validate a digital model of the existing lighting
conditions with which building-wide illuminance values and luminance renderings were generated. With the simulation
method successfully validated, a digital “shoebox” model of a conventionally lit office space was created and an unglazed
baseline condition generated. Various façade conditions were attributed to the model, including films, shades, an insulated
glazing unit (IGU) with louvered blinds, and an integrated glazing unit with a cellular structure developed by Panelite.
Simulation revealed reductions in illuminance by the cellular structure IGU, louvers oriented at 45º, and closed louvers, with
significant reductions achieved with the use of shade cloth. Simulation renderings revealed luminance reduction of 60% by
opened louvered blinds, 94% for louvered blinds oriented at 45º, and 99% for closed louvered blinds. Luminance reductions
for the Panelite structure were estimated at 91% based on simulation results, though visualizations were inconclusive.
Building managers and designers who contend with strict light pollution regulations may accept façade mitigation
simulations as a source for support. By addressing light spill with compatible metrics and validation strategies, these
professionals could make the case for built-in sensitivity to exfiltration. Understanding the role of the façade construction and
the light pollution reduction potential of materials may increase these opportunities for building-specific light management
strategies.

HYPOTHESIS
An Integrated Glazing Unit (IGU) with a cellular structure is able to substantially reduce interior night-lighting losses
through a commercial building façade unit. The physical construction of the IGU is able to drastically reduce light levels by
cutting off light propagation which results in uplight leading directly to sky glow. Furthermore, a cellular IGU structure will
shield light leaving a façade at a downward angle, leading to light trespass. The construction of a multi-pane IGU with a
cellular structure will significantly reduce window luminance values leading to glare conditions for light loss normal to the
façade plane, while maintaining views through the system.

Chow | 4

TABLE OF CONTENTS

Committee......................................................................................................................................................................................2
Abstract ..........................................................................................................................................................................................3
Hypothesis .....................................................................................................................................................................................3
Table of Contents ..........................................................................................................................................................................4
Acknowledgements ........................................................................................................................................................................8
List of Figures ................................................................................................................................................................................9
List of Tables ...............................................................................................................................................................................11
1 INTRODUCTION ............................................................................................................................................................12
1.1 Façade Permeability ..............................................................................................................................................12
1.1.1 Hypothesis Statement .......................................................................................................................................12
1.2 Outside-in / Inside-out ............................................................................................................................................13
1.2.1 Rapidly Brightening Cities .................................................................................................................................13
1.2.2 Value of the Night Sky .......................................................................................................................................14
1.2.3 International Dark-Sky Association ...................................................................................................................15
1.2.4 Road lighting .....................................................................................................................................................15
1.3 Terms .....................................................................................................................................................................16
1.4 Basic Concepts ......................................................................................................................................................18
1.4.1 Infiltration ...........................................................................................................................................................18
1.4.1.1 Thermal and Light Permeation ......................................................................................................................18
1.4.2 Exfiltration ..........................................................................................................................................................18
1.4.2.1 Thermal Loss ................................................................................................................................................18
1.4.2.2 Light Loss ......................................................................................................................................................18
1.4.3 Glazing Transmissivity ......................................................................................................................................19
1.4.4 Light Pollution ....................................................................................................................................................19
1.4.4.1 Sky Glow .......................................................................................................................................................19
1.4.4.2 Light Trespass ..............................................................................................................................................19
1.4.4.3 Glare .............................................................................................................................................................19
1.4.5 Spectral Content of Light ...................................................................................................................................20
1.4.5.1 Light Spectrum Indicators .............................................................................................................................20
1.4.5.2 Human Spectral Sensitivity and the Environment .........................................................................................20
1.4.6 Environmental Impacts of Light Pollution ..........................................................................................................21
1.4.6.1 Nocturnal Mammals ......................................................................................................................................21
1.4.6.2 Birds ..............................................................................................................................................................21
1.5 Study boundaries ...................................................................................................................................................22

Chow | 5

2 RESEARCH REVIEW: Building Contributions to Light Pollution ....................................................................................23
2.1 Light Pollution Perspectives: Methods and Modeling ............................................................................................23
2.1.1 Sky-Down Approach ..........................................................................................................................................23
2.1.2 Ground-Up Approach ........................................................................................................................................28
2.1.3 Comprehensive Studies ....................................................................................................................................28
2.2 Light Propagation: Illumination at the Façade Plane .............................................................................................28
2.3 Building Sourced Light Pollution: Research Interests ............................................................................................32
2.3.1 Circadian Wellness and the Built Environment .................................................................................................32
2.3.2 Bird Window Collisions ......................................................................................................................................33
2.4 Research Synthesis ...............................................................................................................................................36
3 METHODOLOGY ...........................................................................................................................................................38
3.1 Case Study Selection: 800 Wilshire Boulevard .....................................................................................................38
3.2 Phase 1: Data Collection .......................................................................................................................................41
3.2.1 Measurement ....................................................................................................................................................42
3.2.1.1 Survey ...........................................................................................................................................................42
3.2.2 Data Collection Module .....................................................................................................................................43
3.2.3 Photometric Analysis .........................................................................................................................................44
3.2.3.1 Luminance Photographs ...............................................................................................................................44
3.2.3.2 Illuminance Measurements ...........................................................................................................................44
3.3 Phase 2: Data Processing .....................................................................................................................................45
3.3.1 Modeling ............................................................................................................................................................45
3.3.2 Digital Simulation ...............................................................................................................................................46
3.3.2.1 Validation of the Digital Model ......................................................................................................................47
3.4 Phase 3: Reduction Testing ...................................................................................................................................47
3.4.1 Modeling ............................................................................................................................................................47
3.4.2 Digital Simulation ...............................................................................................................................................47
3.4.2.1 Comparison of Reduction Potential ..............................................................................................................48
4 RESULTS .......................................................................................................................................................................50
4.1 Phase 1: Case Study Data Collection....................................................................................................................50
4.1.1 Survey ...............................................................................................................................................................50
4.1.2 Illuminance ........................................................................................................................................................53
4.1.3 South Façade Unshaded ...................................................................................................................................53
4.1.4 South Façade Shaded .......................................................................................................................................54
4.1.5 Luminance .........................................................................................................................................................55
4.1.6 South Façade Unshaded ...................................................................................................................................55
4.1.7 South Façade Shaded .......................................................................................................................................57
Chow | 6

4.2 Phase 2: Digital Simulation ....................................................................................................................................59
4.2.1 Illuminance ........................................................................................................................................................59
4.2.1.1 South Façade Unglazed ...............................................................................................................................59
4.2.1.2 South Façade Unshaded ..............................................................................................................................59
4.2.1.3 South Façade Shaded ..................................................................................................................................60
4.2.2 Luminance Renderings .....................................................................................................................................61
4.2.2.1 South Façade Unshaded ..............................................................................................................................61
4.2.3 South Façade Shaded .......................................................................................................................................62
4.3 Phase 3: Reduction Potential ................................................................................................................................63
4.3.1 Illuminance ........................................................................................................................................................63
4.3.1.1 Unglazed Condition .......................................................................................................................................63
4.3.1.2 Single Pane Clear .........................................................................................................................................64
........................................................................................................................................................................................64
4.3.1.3 Double Pane Clear ........................................................................................................................................65
4.3.1.4 Double Pane Low E ......................................................................................................................................66
4.3.1.5 Triple Pane Krypton Air Gap .........................................................................................................................67
4.3.1.6 SolarGard Film ..............................................................................................................................................68
4.3.1.7 Mechoshade with 5% Openness Factor .......................................................................................................69
4.3.1.8 SolarGard Film and Mechoshade .................................................................................................................70
4.3.1.9 Open Louvered Shade ..................................................................................................................................71
4.3.1.10 Louvered Shade 45º .....................................................................................................................................72
4.3.1.11 Closed Louvered Shade 45º .........................................................................................................................73
4.3.1.12 Cellular Insulated Glazing Unit (IGU) ............................................................................................................74
4.3.2 Luminance .........................................................................................................................................................75
4.3.2.1 Unglazed .......................................................................................................................................................75
4.3.2.2 Single Pane Clear Glass ...............................................................................................................................75
4.3.2.3 Double Pane Clear Glass .............................................................................................................................75
4.3.2.4 Double Pane Low Emissivity Glazing ...........................................................................................................76
4.3.2.5 Triple Pane Krypton Gap ..............................................................................................................................76
4.3.2.6 SolarGard Film ..............................................................................................................................................76
4.3.2.7 Mechoshade 5% Openness Factor...............................................................................................................77
4.3.2.8 SolarGard Film and Mechoshade .................................................................................................................77
4.3.2.9 Louvered Blind ..............................................................................................................................................78
4.3.2.10 Cellular Insulated Glazing Unit ......................................................................................................................79

Chow | 7

5 DISCUSSION .................................................................................................................................................................80
5.1 Discusison of Data Collection ................................................................................................................................80
5.1.1 Case Study Considerations ...............................................................................................................................80
5.1.2 Survey ...............................................................................................................................................................80
5.1.3 Data Collection Module .....................................................................................................................................80
5.1.4 Photometric Analysis .........................................................................................................................................81
5.1.4.1 Illuminance ....................................................................................................................................................81
5.1.4.2 Luminance ....................................................................................................................................................82
5.2 Discussion of Data Processing ..............................................................................................................................82
5.2.1 Digital Modeling Considerations ........................................................................................................................82
5.2.1.1 Variables in Rhinoceros ................................................................................................................................82
5.2.1.2 Variables in DIVA ..........................................................................................................................................83
5.2.2 Digital Model Validation .....................................................................................................................................85
5.2.2.1 South Façade Unshaded ..............................................................................................................................86
5.2.2.2 South Façade Shaded ..................................................................................................................................89
5.3 Discussion of Reduction Potential .........................................................................................................................91
5.3.1 Digital Modeling Considerations ........................................................................................................................91
5.3.1.1 Variables in Rhinoceros ................................................................................................................................91
5.3.1.2 Variables in DIVA ..........................................................................................................................................91
5.3.2 illuminance.........................................................................................................................................................94
5.3.2.1 Illuminance Maps ..........................................................................................................................................95
5.3.3 luminance ..........................................................................................................................................................99
5.3.4 Renderings ........................................................................................................................................................99
6 CONCLUSIONS ...........................................................................................................................................................101
6.1 Conclusion of Findings ........................................................................................................................................101
6.1.1 Window Illuminance ........................................................................................................................................101
6.1.2 Interior Luminance ...........................................................................................................................................101
6.1.3 Direct Sky Glow ...............................................................................................................................................101
6.1.4 Research Conclusions ....................................................................................................................................101
6.2 Application of Research .......................................................................................................................................102
6.2.1 Changing Regulations .....................................................................................................................................102
6.2.2 Invested Professionals ....................................................................................................................................102
6.2.3 Whole Building Analysis ..................................................................................................................................103
6.2.4 Value of the (Night) Skyline .............................................................................................................................104
6.3 Concluding Remarks ...........................................................................................................................................104
BIBLIOGRAPHY ........................................................................................................................................................................105
Chow | 8

ACKNOWLEDGEMENTS
I am honored to recognize the combined efforts of the professionals who endeavored to build this work. I offer my immense
gratitude to the members of my Thesis Committee; Dr. Douglass Noble, Dr. Travis Longcore, and Marc Schiler, who offered
their scrutiny, ingenuity, and expertise in three seemingly disparate specialties in order to support the topic of this research.
I am especially indebted to my Committee Chair, Dr. Douglas Noble, who challenged me and championed my place in the
program.
I am happy to pay special recognition to the venerable researchers, building scientists, mentors, and colleagues in the fields of
lighting design, astrophysics, and building engineering, who expressed interest and confidence in this work.
I am immensely grateful to my Building Science family, which continues to grow around the world from indelible roots in
the University of Southern California. There is no greater group of young professionals that can be relied upon to build a
better world.
Finally, I acknowledge the full support of my family and loved ones. Thank you for your inspiration and borrowed strength.

Chow | 9

LIST OF FIGURES
Figure 1.1: Increases in U.S. Night Sky Brightness; Adapted from Elvidge Cinzano, Falchi (2001); Diagram by Author ...........13
Figure 1.2: Earth's City Lights as seen via Defense Meteorological Satellite Program (DMSP) Operational Linescan System
(OLS). Data courtesy Marc Imhoff of NASA GSFC and Christopher Elvidge of NOAA NGDC. Image by Craig Mayhew and
Robert Simmon, NASA GSFC. ....................................................................................................................................................14
Figure 1.3: Spectral Power Distribution Curves for Two Sources; Cool White Fluorescent (Left) and Incandescent (Right) Light
Sources (lrc.rpi.edu) ....................................................................................................................................................................20
Figure 1.4: Estimates of Annual Bird Mortality by Collisions with U.S. Buildings (Loss et al., 2014) ..........................................22
Figure 2.1: Flowchart Linking Sky Glow to Artificial Light Use (Luginbuhl et al.,2009)................................................................23
Figure 2.2: Upward (Left) and Downward (right) artificial radiation density Tb km
3
(Cinzano et al, 2013) ..................................25
Figure 2.3: VIIRS/DNB Images over Madrid for April 2012, October 2012, January 2013 and May 2014(Right to left, Top to
Bottom); Color Scale Indicates Radiance in nW/cm
2
/sr (Sanchez de Miguel, 2015)...................................................................27
Figure 2.4: VIIRS/DNB Observations for Pronounced Viewing Angles May Be Obscured by Buildings Taller Than 6 Stories
(18m) (Sanchez de Miguel, 2015) ...............................................................................................................................................27
Figure 2.5: Celestial Hemisphere at Point P as Captured by CCD Camera with Fisheye Lens Showing Lighting Sources From
Outdoor Lighting and Building Windows (Oba et al. 2005) ..........................................................................................................29
Figure 2.6: Changing Illuminance Levels from Outdoor Lighting Sources over Time for Two Sources (Oba et al. 2005) ..........29
Figure 2.7: Sample Building Survey ............................................................................................................................................30
Figure 2.8: Light Propagation Patterns through Windows (Darula et al., 2013) ..........................................................................31
Figure 2.9: Idealized Effects of Various Shading Devices on Interior Light Loss Trajectories (Darula et al. 2013) ....................32
Figure 2.10: Pixel Analysis of Illuminated Window Counts from a Single Vantage Point (Parkins, Elbin, and Barnes, 2015) ....33
Figure 2.11: Photos of the East Side Windows without Exterior Shades (Left) and with Exterior Shades (Right), .....................36
Figure 3.1: Three-Phase Methodology ........................................................................................................................................38
Figure 3.2: Location of Case study Building at 800 Wilshire Blvd, Los Angeles, Google Maps 2015 .........................................40
Figure 3.3: Buro Happold Offices on the 16
th
Floor of 800 Wilshire; Open Office (Left), and Balcony (Right) ............................40
Figure 3.4: Diagrammatic Summary of Phase 1 Data Collection Procedure. Note the Survey and Photometric Analysis
Required Minimal Software Processing. ......................................................................................................................................41
Figure 3.5: Data Collection Module Design and Critical Dimensions. .........................................................................................43
Figure 3.6: Data Collection Module Erected in Place at the Case study Façade Plane. Note the Locations of Equidistant Test
Nodes within the Module. ............................................................................................................................................................43
Figure 3.7: Composite Photograph of Data Collection Procedure ...............................................................................................44
Figure 3.8: Diagrammatic Summary of Phase 2: Data Processing Simulation Procedures Note: Relevant Finish and Lighting
Strategy Information from the Initial Building Survey ...................................................................................................................45
Figure 3.9: Screenshot of the Digital Model Replicating the 16
th
Floor Terrace. .........................................................................45
Figure 3.10: Calculation Node Locations for Data Processing Digital Model, Note the Intermediate Nodes and Equidistant
Point Spacing ...............................................................................................................................................................................46
Figure 3.11: Diagrammatic Summary of Phase 2: Data Processing Simulation Procedures Note: Relevant Finish and Lighting
Strategy Information from the Initial Building Survey ...................................................................................................................47
Figure 3.12: Screenshot of the Digital Model, Note the Layers Generated by Material ..............................................................47
Figure 3.13: Screenshot of the Digital Model Describing a Glazed Condition for Reduction Testing .........................................48
Figure 4.1: Diagrammatic Results for the Visual Survey .............................................................................................................50
Figure 4.2: Critical Façade Dimensions and Considerations Required for the Digital Modeling Process ...................................52
Figure 4.3 South Façade, Unshaded Bracketed Images taken with a Nikon D300.....................................................................55
Figure 4.4: Physical Data Collection; So. Facade Unshaded HDR Composite Photo ................................................................55
Figure 4.5: Physical Data Collection; So. Facade Unshaded False Color Luminance Image; values in candela/m
2
..................56
Figure 4.6: Physical Data Collection; South Facade Shaded, Bracketed Images taken with a Nikon D300 ...............................57
Figure 4.7: Physical Data Collection; So. Facade Shaded HDR Composite Photograph ...........................................................57
Figure 4.8: Physical Data Collection; So. Facade Shaded False Color Luminance Image, Values in candela/m
2
.....................58
Chow | 10

Figure 4.9: Digital Simulation; So. Facade Unshaded False Color Luminance Rendering, Scale in candela/m
2
.......................61
Figure 4.10: Digital Simulation; So. Facade Shaded Luminance Rendering, Scale in candela/m
2
.............................................62
Figure 4.11: Reduction Potential; Luminance Rendering of Unglazed Base Case, Scale in candela/m
2
...................................75
Figure 4.12: Reduction Potential; Luminance Rendering of Single Pane Clear Glass, Scale in candela/m
2
..............................75
Figure 4.13: Reduction Potential; Luminance Rendering, Double Pane Clear Glass, Scale in candela/m
2
................................75
Figure 4.14: Reduction Potential; Luminance Rendering, Double Pane Low E, Scale in candela/m
2
........................................76
Figure 4.15: Reduction Potential; Luminance Rendering, Triple Pane Glazing with Air Gap, Scale in candela/m
2
....................76
Figure 4.16: Reduction Potential; Luminance Rendering, Clear Glazing with Solar Control Film, Scale in candela/m
2
.............76
Figure 4.17: Reduction Potential; Luminance Rendering, Clear Glazing with 5% Open Shade Cloth, Scale in candela/m
2
......77
Figure 4.18: Reduction Potential; Luminance Rendering, Solar Control Film and Mechoshade Shade Cloth, Scale in
candela/m
2
...................................................................................................................................................................................77
Figure 4.19: Reduction Potential; Luminance Rendering, Double Glazing 76mm Louvered Shade Open .................................78
Figure 4.20: Reduction Potential; Luminance Rendering, Double Glazing 76mm Louvered Shade 45° Orientation ................78
Figure 4.21: Reduction Potential; Luminance Rendering, Double Glazing 76mm Louvered Shade Closed ...............................78
Figure 4.22: Reduction Potential; Luminance Rendering, Double Glazing 12mm White Cellular Structure ...............................79
Figure 5.1: Diagram Showing the Data Collection Module at the Testing Façade, Note the non-Contributing Portion of the
Façade Unit Shaded in Blue ........................................................................................................................................................81
Figure 5.2: Location of Data Points within Data Collection Module .............................................................................................82
Figure 5.3: Rhinoceros 5.0 DIVA Toolbar ....................................................................................................................................83
Figure 5.4: Horizontal Plane Calculation Nodes ..........................................................................................................................83
Figure 5.5: South Façade Unshaded Condition HDR Compiled Image (Left) and Rendered (Right) Perspective Views from 1m
.....................................................................................................................................................................................................86
Figure 5.6: Diagrammatic Comparison of Physical (Left) and Digital Illuminance Measurements (Right) ..................................86
Figure 5.7: South Facade Unshaded Condition HDR False Color (Left) and Rendered (Right) Luminance Maps .....................88
Figure 5.8: South Facade Shaded Condition HDR False Color (Left) and Rendered (Right) Luminance Maps ........................90
Figure 5.9: Digital Shoebox Model Reflecting Conventional Offices at 800 Wilshire ..................................................................91
Figure 5.10: WINDOW7.4 Interface; Panelite Clearshade Optical Data Results ........................................................................93
Figure 5.11: Average Illuminance Values for Reduction Testing Conditions by Façade Treatment, Values in lux .....................94
Figure 5.12: Reduction Potential Illuminance Maps for Unglazed Condition (Left) and various DIVA Glazing Materials ...........95
Figure 5.13: Reduction Potential Illuminance Maps for Unglazed Condition (Left), Solargard Film and Mechoshade Shade
Cloth (Right) .................................................................................................................................................................................96
Figure 5.14: Reduction Potential Illuminance Maps for Unglazed Condition (Left), Louvered Blinds Opened, Oriented at 45º,
and Closed (Right) .......................................................................................................................................................................97
Figure 5.15: Reduction Potential Illuminance Maps for Unglazed Condition (Left), Insulated Glazing Unit with Cellular Structure
(Right) ..........................................................................................................................................................................................98
Figure 5.16: Luminance Reduction at 1m from the Façade by Shading Type ............................................................................99
Figure 5.17: Cellular Structure Visualization, Transmissivity Simulation (Left), Expected Visualization (Middle), Modeled
Visualization (Right); image credit: Panelite.us .........................................................................................................................100

Chow | 11

LIST OF TABLES
Table 2.1: Number and Description of Buildings in Each Broad Class Used in Simulation (Machtans, Wedeles, and Bayne
2013) ............................................................................................................................................................................................34
Table 2.2: Summary of Estimates of Bird Mortality Caused by Bird-Window Collisions at Different Types of Buildings in
Canada (Machtans, Wedeles, and Bayne 2013) .........................................................................................................................35
Table 3.1: Important Case study Building Survey Considerations ..............................................................................................42
Table 3.2: Case study Conditions Survey....................................................................................................................................42
Table 4.1: Physical Data Collection; Lighting Survey Results for 16th Floor Open Offices.........................................................51
Table 4.2: Physical Data Collection; Material Survey for 16th Floor Open Offices .....................................................................51
Table 4.3: Physical Data Collection; Shading Survey for 16
th
Floor Open Offices ......................................................................51
Table 4.4: Physical Data Collection; Illuminance Measurements, South Facade Unshaded ......................................................53
Table 4.5: Physical Data Collection; Illuminance Measurements, South Facade Shaded ..........................................................54
Table 4.6: Data Processing; Illuminance Measurements South Facade Unglazed, Values expressed in lux ............................59
Table 4.7: Data Processing; Illuminance Measurements South Facade Unshaded ...................................................................59
Table 4.8: Data Processing; Illuminance Measurements South Facade Unshaded ...................................................................60
Table 4.9: Reduction Potential; Unglazed Illuminance Calculations ...........................................................................................63
Table 4.10: Reduction Potential; Single Pane Clear Illuminance Calculations............................................................................64
Table 4.11: Reduction Potential; Double Pane Clear Illuminance Calculations ..........................................................................65
Table 4.12: Reduction Potential; Double Pane Low E Illuminance Calculations .........................................................................66
Table 4.13: Reduction Potential; Triple Pane Krypton Gap Illuminance Calculations .................................................................67
Table 4.14: Reduction Potential; SolarGard Film Illuminance Calculations ................................................................................68
Table 4.15: Reduction Potential; Mechoshade Illuminance Calculations ....................................................................................69
Table 4.16: Reduction Potential; Shade and Film Illuminance Calculations ...............................................................................70
Table 4.17: Reduction Potential; Louvered Open Shade Illuminance Calculations ....................................................................71
Table 4.18: Reduction Potential; Louvered 45 Deg. Shade Illuminance Calculations.................................................................72
Table 4.19: Reduction Potential; Louvered Closed Shade Illuminance Calculations ..................................................................73
Table 4.20: Reduction Potential; Cellular IGU Illuminance Calculations .....................................................................................74
Table 5.1: Physical Data Collection; Illuminance Measurements South Facade Unshaded .......................................................82
Table 5.2: Custom DIVA Materials ..............................................................................................................................................84
Table 5.3: South Façade Unshaded Illuminance Measurements, Estimated Accuracy ..............................................................87
Table 5.4: South Façade Unshaded Illuminance Measurements (Left) Calculations (Right) With Reduced Furniture
Reflectance Values, Note Significant Reductions in Simulated Values (Right) for Furniture Light Reflectance Values .............88
Table 5.5: South Façade Shaded; Illuminance Measurements (Left) Calculations (Right) .........................................................89
Table 5.6: Reduction Potential; Custom Radiance Materials for use in DIVA .............................................................................92

Chow | 12

1 INTRODUCTION
The night presence of buildings helps create iconic urban identities for densely populated cities across the world. Though
beautiful, a visible skyline draws attention to potentially significant light losses from building interiors. Accounting for
variable lighting strategies and patterns of illumination, it is unclear how much light is escaping through a building façade
after dark. What is very clear, is that there is an excess of light in the urban environment. The presence of excess light in the
night sky as light pollution is a problem currently being researched to great lengths by professionals in the natural and
atmospheric sciences, medicine, ecology, astronomy, and landscape and lighting design. Despite the cross-disciplinary
interest of researchers, advocacy groups, and government entities committed to protecting the night sky from the brightening
consequence of rapidly-developing cities, little research exists to quantify interior night light exfiltration through glazed
facades. The potentially significant impact of reducing these light losses is ample motivation to understand a building’s
contribution to the effects of light pollution. Considering these multi-disciplinary interests, commercial buildings were
evaluated as contributors of light pollution and potential strategies for mitigating this wide-reaching issue.
1.1 FAÇADE PERMEABILITY
Various types of construction are defined broadly by the term “façade.” Facade diversity has been profoundly influenced by
both utilitarian and expressionistic needs of human beings across the world’s cultures and climates. Humanity’s need for
shelter and ensuing desire to develop its aesthetics during the evolution of the architectural discipline have produced
enclosures which range from structural to ephemeral. Façade design and engineering continue to expand, and while
vernacular shelters retain their validity amidst new materials and methods, the primary role of the building façade remains the
same: to mitigate exterior conditions to promote comfortable interior conditions. The building envelope, as it is often called
in the architecture, engineering, and construction industries, connotes a relationship between container and content that is
managed by a seal. While appropriate in some ways, this image misrepresents the reality of façade permeability and its
influence on façade design. It is the duty of the architect, engineer, or façade consultant to regulate the permeability of this
seal to allow appropriate light infiltration and protect the contents of a building from the potentially damaging consequences
of daylight. After dark, however, an interior-lit building engages in an inverse relationship with the exterior environment,
disrupting natural patterns of light and dark for plant and animal life (see Section 1.4.6).
1.1.1 HYPOTHESIS STATEMENT
Commercial building façade permeability may be studied for its role in contributing to forms of light pollution not sourced by
road lighting. A dedicated data capture and simulation methodology can be employed to accurately evaluate light loss
through a building façade for use in sky glow, glare, and light trespass management. A digital simulation process can
accurately be used to illustrate the reduction potential of various façade constructions which may act as mitigation strategies
for building-sourced light pollution. Shading strategies may be evaluated at the façade unit scale, using luminance renderings
to measure reductions in glare sources and illuminance calculations to quantify light trespass and direct uplight.

Chow | 13

1.2 OUTSIDE-IN / INSIDE-OUT
Understanding light passage through a façade is critical for the design of contemporary buildings. Daylight infiltration
through a façade is important to building professionals across various disciplines as well as building owners, operators and
occupants. The associated costs and efforts of thinking “outside/in” have been extensively studied (Aksamija 2013).
Reversing the trajectory of conventional thinking, the “inside/out” relationship between light and a building façade suggests a
fascinating array opportunities for research. Considering seventy percent of a town’s luminous flux can be attributed to
urban roadway and parking lot lighting, building designers and engineers must become aware and capable of designing for a
building’s possible contribution to the remaining thirty percent of excess light in the urban environment (Soardo et al. 2008).
The presence of excess light in the night sky as light pollution is a problem currently being researched to great lengths by
professionals in medicine, ecology, astronomy, and landscape and lighting design. There has never been more motivation to
understand a building’s contribution to the effects of light pollution. Considering these multi-disciplinary interests,
commercial buildings were evaluated as contributors of light pollution along with potential strategies for mitigating this
wide-reaching issue. Interior night lighting losses through a commercial building façade were investigated to examine the
influence of light exfiltration on some of the planet’s gravest concerns.
1.2.1 RAPIDLY BRIGHTENING CITIES
In 1998, Cinzano and Falchi published their report of artificial night sky brightness in Italy taken from NOAA Defense
Meteorological Satellite Program (DMSP) measurements. Their research continues the night lighting maps generated by
researchers Walker (1970, 1973) and Albers (1998) in the USA, Bertiau, de Graeve, and Treanor (1973; Treanor 1974) in
Italy, and Berry (1976) in Canada, who generated light pollution estimates from population data (Cinzano et al. 1999). By
comparing their maps with data sourced by Bertiau et al. (1973), Cinzano and Falchi generated a 25-year growth profile for
the country. Based on the artificial sky brightness increase, the researches extrapolated a similar map for the year 2025,
which suggests a sky brightness increase to over 100 times the natural sky brightness. In 2001, the same researchers
published the first world atlas of artificial night sky brightness (Cinzano et al. 2001). This global view of artificial night sky
brightness at sea level attributes lighting density to the planet’s urban sources at a given time. While referencing the United
States Department of Energy population density data, the atlas reveals a determinate number of people living under a section
of sky of a given brightness (Cinzano, Falchi, and Elvidge 2001). By assembling similar snapshots of urban light sources, as
they had done for the various regions in Italy, Cinzano and Falchi were able to map and predict growing night sky brightness
for specific urban areas. Figure 1.1 illustrates their predictions for sky brightness increases in North America, adapted from
their findings.

Using a similar logic, NASA researcher Marc Imhoff worked closely with Cristopher Elvidge to tie surges in night sky
brightness to city growth and the effects of urbanization. Figure 1.2 illustrates a global view of urban brightness. Using the
same Operational Linescan System (OLS) of the DMSP network, Imhoff and his team investigated the spatial characteristics
of urbanized areas independently of population and confirmed what the Italian researchers helped predict: Cities and their
night-lighting footprints are growing (Weier 2000).
Figure 1.1: Increases in U.S. Night Sky Brightness; Adapted from Elvidge Cinzano, Falchi (2001); Diagram by Author
Chow | 14

Imhoff was inspired with night-time satellite imaging, which had been used by past census researchers to estimate population
density in urban areas. Incorporating other sources of data, these images carry rich information for researchers of various
disciplines. Independent of population, these images have been used to extrapolate clues for understanding economic
disparity, population behavior, and citizen well-being (Ghosh et al. 2013). Several significant studies have utilized the Dark-
Sky-friendly city of Flagstaff, Arizona to tie spatial and building age-related factors to trends in increasing artificial night
lighting (Lockwood, Floyd, and Thompson 1990). In another study, night light images taken by a specialty aircraft over
Berlin, Germany were paired seamlessly with ground-based land use data forming a high resolution light pollution map with
the potential to identify spatial light trends at a precise scale. The analysis of mosaic imaging with a 1-m resolution and land
use maps helped researchers categorize sources of light pollution into distinct percentages of an overall study area. These
light emission percentages described the high and low emission areas such as streets and parks, and grouped different
building types identified by land based surveys. The research team differentiated public service structures such as hospitals
and schools from high rise buildings and single-family residences, and even categorized the emissions from the courtyards of
“block buildings” typical of Berlin. Helga U. Kuechly and her team demonstrated vast differences between this dataset and
other similarly investigated locations, citing various opportunities for the research to facilitate the regulation of artificial
lighting and tracking the success of regulatory programs (Kuechly et al. 2012).

Figure 1.2: Earth's City Lights as seen via Defense Meteorological Satellite Program (DMSP) Operational Linescan System (OLS). Data
courtesy Marc Imhoff of NASA GSFC and Christopher Elvidge of NOAA NGDC. Image by Craig Mayhew and Robert Simmon, NASA
GSFC.
1.2.2 VALUE OF THE NIGHT SKY
The value of the night sky has been argued by researchers interested in the sociological impacts of light pollution. Many
agree that the night sky is a highly valuable commodity which is being lost due to the decreasing contrast of faint stars by the
veil of luminous energy scattered in the atmosphere (Boyce 2014). In addition to supporting any practical motivations for
maintaining a clearly visible sky, several nostalgic researchers regret the loss of “Nature’s grandest free show,” which has
inspired humanity’s curiosity and core religious and cultural themes (Mizon 2012). Researchers believe this rapid alteration
to the natural environment and the inevitable loss of perception of the Milky Way’s natural beauty may have unintended
societal consequences (Cinzano, Falchi, and Elvidge 2001). To help educate the public and guard against these unforeseen
consequences, global advocacy groups and organizations have developed innovative programs and influential stewardship
initiatives (Parks 2014).

Chow | 15

1.2.3 INTERNATIONAL DARK-SKY ASSOCIATION
Of the many organizations committed to the reduction of light pollution, the International Dark-Sky Association (IDA) has
perhaps the largest influence over public awareness of the issue. The IDA is primarily concerned with stewardship of the
night sky, educating the public on the subject of light pollution and its damaging effects to the planet’s fragile ecosystems.
The IDA provides authoritative, third-party consultation for lighting industry leaders, policy makers, and conservation
organizations. The IDA also certifies fixtures with an IDA Fixture Seal of Approval, designating outdoor luminaires which
minimize glare and reduce light trespass and light pollution. Protection programs designed by the IDA are aimed at
maintaining a high quality of sky visibility in parks, natural preserves, and similar outdoor environments. A strict certification
process is required for certification as a protected place (“International Dark-Sky Association” 2015)
The IDA’s 2014 Smart Urban Lighting Initiative stresses simple goals to reduce carbon emissions, energy costs, and light
pollution in cities which have embraced solid state lighting (SSL) technology. Through this initiative, the IDA has worked
closely with lighting manufactures, engineers and designers, and public safety professionals to provide quality light and
optimal outdoor lighting strategies for interested communities (Parks 2014).
1.2.4 ROAD LIGHTING
In terms of light pollution, much of the debate in the reduction of urban street lighting has been silenced by issues of safety
and economy. However, efforts to capture energy savings from roadway lighting may have a positive effect on light
pollution. In an effort to capture energy savings, planning officials at the community level have spear-headed campaigns to
retrofit their existing street and road lighting. Outfitted with various control systems, roadway luminaires are capable of being
dimmed in the absence of vehicles or pedestrians. The diming potential of these fixtures may prove to be the maximum
reduction available to roadway lighting before jeopardizing safety or security.

Chow | 16

1.3 TERMS
This study is concerned with various interrelated disciplines, therefore, it is essential to define some fundamental terms that
arise.
Façade Infiltration
The mitigated or unmitigated passage of light or heat energy from the exterior environment, through a building
facade to the building interior.
Commercial Buildings
As defined by the California Energy Commission; a building whose primary function is non-residential. These
buildings do not have 24-hour occupancy. Examples of commercial buildings include office towers, and non-mall
retail applications.
Façade Light Exfiltration; Light Loss, Light Leak
These terms, used interchangeably, refer to the intended or coincidental passage of light energy from the interior of a
building, through its facade to the outdoor environment.
Light Pollution
Light Pollution refers to the propagation of direct or indirect light to the atmosphere by manmade sources, which
contributes to sky glow, glare, and light trespass.
In astrophysics, the alteration of the natural quantity of light in the night environment due to introduction of
manmade light (Cinzano et al, 2013).
In the ecological sciences, the distinction has been made between Astronomical and Ecological effects of light
pollution (Longcore and Rich 2004). Astronomical Light Pollution refers to the broad-scale veiling effect of
cumulative light in the environment on the perception of celestial bodies. Ecological light pollution is used to
describe the artificial light which alters the natural patters of light and dark in ecosystems, and may be used
interchangeably with Verheijen’s photopollution, meaning “light having adverse effects on wildlife,” in the section
concerning this topic (Longcore and Rich 2004).
Sky Glow
Refers to the increase in the luminance of the sky at night above that produced by natural sources such as
moonlight (Boyce 2014)
Glare
A physiological and psychological phenomena caused by light scattered in the eye and by a range of
luminance present in the visual field, psychologically, it is related to the experience of reduced visibility as
well as more general discomfort and irritation
Light Trespass
A local phenomenon in which significant amounts of light from nearby luminaires cross a property
boundary, causing disturbances to an adjacent owner’s ability to enjoy the use of his/her property
Lighting Engineering Indicators
The measurements described below are almost always defined by the sensitivity of the human eye, adapted to light
conditions of the observer’s environment.
Luminous Flux
Describes the total luminous energy leaving a theoretical surface of one square meter at a distance of one
meter (Schiler 1992). Luminous flux is expressed by lumens (lm) or Talbots per second (T/s).
Luminous Intensity
Refers to the amount of light emitted by a light source in a particular direction (Schiler 1992). The SI unit
for luminous intensity is the candela (cd).
Illuminance
Defines the luminous flux density arriving at a real surface (Schiler 1992). The SI unit for illuminance is
lux (lx).

Chow | 17

Luminance
References the total amount of luminous flux density leaving a projected surface in a given direction
(Schiler 1992). The SI unit for luminance is the candela per square meter, (cd/m²).

Chow | 18

1.4 BASIC CONCEPTS
In commercial applications, the facade construction becomes a barrier between dynamic exterior conditions and a sustained
sense of occupant comfort within. Light has a profound effect on buildings as a primary source of flux. Early energy
modeling can provide benchmarks around which to design appropriate mechanical systems using known data about the
façade, and deliver strategies for regulating the inevitable infiltration of daylight-driven forces. These forces contribute
greatly to the design and operation of contemporary mechanical systems and therefore, the occupant comfort for which they
have become necessitated. Put simply, recognizing light as energy is critical to understanding the importance of infiltration
and ultimately, exfiltration.
1.4.1 INFILTRATION
1.4.1.1 THERMAL AND LIGHT PERMEATION
Façade permeability is most commonly studied in terms of infiltration, by which daylighting enters the glazed area of a
building façade, heating the interior by radiative, convective, conductive means. In the case of commercial new construction,
the façade assembly is placed under simulated scrutiny and becomes a major factor in the estimation of electrical loads
necessary for maintaining occupant comfort. Much of the consideration for early energy modeling requires information
regarding the façade construction, material, and glazing surface area to simulate the infiltration of exterior forces to the
interior. The resulting estimation of Energy Use Intensity (EUI) is then used to appropriately select a mechanical system
capable of maintaining a level of indoor environmental quality in response to the predicted loads. With so much at stake, it is
no wonder building professionals are concerned with infiltration through the facade. Its measurable influence on costly
energy-dependent systems has perhaps prevented equal consideration of façade exfiltration. The inverse ability for the façade
to allow light to escape a building interior is becoming more relevant as more research is being conducted in the realm of
light pollution and health.
1.4.2 EXFILTRATION
1.4.2.1 THERMAL LOSS
The process of exfiltration has been extensively studied in regards to thermal inefficiency. Differences in exterior and
interior air pressure are facilitated by a project’s HVAC system, and cause air movement to test a façade’s ability to regulate
these forces. Because of the limitations of “real world” construction, designers must design for an inevitable degree of
exfiltration (Boswell 2014). The degree by which an allowable minimum is maintained is determined by material selection
and the efficient application of these materials.
Foresight and an understanding of the design process can avoid large levels of energy loss through an opaque enclosure. The
recent design of the University of Maryland’s IT/E Building promises to put an end to exfiltration with the help of an
innovative cladding approach. The designers questioned traditional construction processes and attempted to preemptively
control the permeability of the façade construction. The traditional method of installing a vapor barrier on the interior of a
building construction leaves it vulnerable to breaches by following trades. For example, if moisture and condensation enter
the construction, connections between studs and screws can be weakened. To avoid compromising the airtightness of the
façade in this project, the air and water barrier is adhered directly to gypsum sheathing and extruded polystyrene insulation.
The construction is located in front of the stud, which frees the stud cavity for the installation of utility lines as needed
(Vinkler 2002). This case study provides a designed response to the threats of exfiltration in a climate where the results can
be very damaging. The careful planning of materials and their ability to resist heat transfer is a concern for building
scientists. Contemporary builders and building scientists continue to innovate their methods to minimize the risk of energy
loss or damage to the façade because of thermal inefficiencies.
1.4.2.2 LIGHT LOSS
Despite the attention paid to exfiltration of heat energy though a building membrane, few data sources document the loss of
light energy or its possible consequences. The excess of artificial light in the night sky is of particular interest to those who
recognize it as a type of environmental pollution. The study of light pollution has gathered significant evidence over the past
25 years which analogizes unmanaged light to other environmental pollutants (Cinzano and Falchi 2012).
In urban areas, the exfiltration of light is oftentimes highly visible. Iconic architecture made to have a night presence by
illuminating sparsely–populated interior spaces. With this “lantern–like” function, commercial buildings contribute to light
Chow | 19

pollution and its harmful effects. While many concerns regarding light pollution are reserved for light which is escaping
vertically into the higher atmosphere, the problem of light loss can me localized to the urban scale. Terrestrial light intrusion
at a shallow angle above the horizon will cause considerably more sky glow. Wasted light up to ten degrees above the
horizon of a luminaire will encounter more particles and droplets through which light may be scattered (Boyce 2014).
1.4.3 GLAZING TRANSMISSIVITY
The infiltration of daylight through a glazing system contributes directly to a designer’s consideration of heat gains and
occupant comfort. Heat gains are highest along a building plan’s periphery, where a building’s glazed façade must mitigate
the solar spectrum. In conventional curtain wall design, the ability for a glazing material to reflect, absorb, and transmit
radiation is given in a ratio of reflectance + absorption + transmittance = 1 (Memari, Architectural Engineering Institute.
Committee on Curtain Wall Systems 2013). These properties, along with emissivity, which describes a material’s ability to
re-radiate long-wave infrared radiation, are unitless ratios between 0 and 1. Frequently used low-emissivity (low-e) metallic
coatings are successful at reflecting and absorbing wavelengths while allowing for visible light transmittance. For the
purpose of night-lighting loss measurements, a glazing system’s potential to reflect and transmit interior light is worthy of
consideration as the degree of transmissivity gives rise to the negative effects of light exfiltration.
1.4.4 LIGHT POLLUTION
The consideration of light as a form of pollution has gained popularity in the last two decades as the effects of night lighting
in urban areas have become impossible to ignore. The overall effect of light pollution threatens the ability to clearly view the
night sky. The diffusion of light by aerosols creates a luminous layer very near to the thresholds of night sky luminance
contrast, reducing the visibility of stars and other astronomical features (Boyce 2014). This visible consequence of light
pollution is experienced globally. Using radiance-calibrated high-resolution DMSP satellite imaging and United States DOE
population density data, researchers have estimated two thirds of the US population in all 50 states have lost the ability to see
the Milky-Way with the naked eye (Cinzano and Falchi 2014). Light pollution is most commonly caused by sky glow, glare,
or light trespass, differentiated by their varying visible effects (Boyce 2014).
1.4.4.1 SKY GLOW
Sky glow is categorized by the increase in the luminance of the sky after dark caused by lighting due to human activity. As
light escapes into the atmosphere, it is scattered by dust particles and suspended water droplets called aerosols. Low levels of
sky glow are common in areas with few aerosols, often characterized by small populations, dry/thin air, and negligible air
pollution (Boyce, 2014). Understandably, more severe instances of sky glow are expected in urban areas where admittedly,
air pollution contributes to higher levels of particles and aerosols. Urban sky glow is most visible as a glowing, flattened
dome of light over cities or towns, where various sources contribute to a cumulative effect. These sources need not be
oriented vertically to the sky. When light hits a surface, a portion of the light is reflected. The amount of light added to sky
glow is dependent on the reflectiveness of the illuminated surface. Up to 70% percent of the luminous flux of an urban area
can be attributed to road lighting, which is reflected by horizontal surfaces (Soardo et al. 2008). It is unclear what effect light
loss through commercial building façades has on the remaining 30% of light perceived.
1.4.4.2 LIGHT TRESPASS
Light trespass occurs when light escaping a luminaire crosses a property boundary, causing disturbances to the owner’s
ability to enjoy the use of his/her property (Boyce 2014). The intrusion of considerable amounts of light can become a source
of annoyance for property owners, and complaints are most commonly in regards to neighbors’ ill-aimed night lighting
encroaching through bedroom windows. Light sources which contribute to light trespass can be varied. In many cases,
nearby light sources are simply unshielded, allowing light to spill in unwanted directions.
1.4.4.3 GLARE
Glare categorizes a physiological and psychological phenomenon at the observer level, commonly described in terms of
disability or discomfort glare. Disability Glare has a direct effect on the visible capabilities of an observer, and can be
quantified by reactions and sensations to physical stimuli. Discomfort glare, is characterized by visible discomfort in the
presence of bright luminaires or windows. Any scatter from light sources may or may not noticeably reduce visibility and
may cause annoyance (Boyce 2014). Since glare is related to the relative ratio of source and background, the issue becomes
more complex in nighttime conditions.
Chow | 20

1.4.5 SPECTRAL CONTENT OF LIGHT
The presence of artificial light in the nighttime environment has been studied for its many influences on the human circadian
system. Sensitivity to the wavelengths of light is considered an evolutionary development which pairs perennial day and
night variations in the environment to biological functions. The complex relationship between human circadian rhythm and
lighting is often understood by examining the spectral content of environment light. Spectral power distribution (SPD) curves
represent the radiant power emitted by a light source between the visible wavelengths of the electromagnetic spectrum
(DiLaura and Illuminating Engineering Society of North America 2011). SPD curves can describe distinct variations between
sources, as illustrated in the SPD curves for fluorescent and incandescent lamps (Figure 1.3). It is important to note the spikes
in radiant power at 400nm and 450nm for fluorescent sources, compared with the relatively low values at the same
wavelengths emitted by incandescent lamps. This graphic method of indicating color output from short (violet) to long (red)
wavelengths is important for understanding the color of a particular source’s output (Panda and Marks 2015).
1.4.5.1 LIGHT SPECTRUM INDICATORS
Correlated Color Temperature (CCT) and Color Rendering Index (CRI) values are also used to describe the “color” of white
light. CCT Values are expressed in Kelvin (K), valuating warm sources at low color temperatures and cool sources at higher
color temperatures (Panda and Marks 2015). A CRI value on a scale of 1-100 is used to describe a lamp’s ability to
accurately render colors in a lighted space (Panda and Marks 2015).The use of these metrics in lighting engineering has been
described as “crude,” as they refer to a hypothetical black body comparison scenario (Aubé, Roby, and Kocifaj 2013). The
CRI scale has been criticized in recent years, leading to the development of an improved color metric by the IES (Wright
2015). Even with this proposed method for describing color rendering, these values are used extensively in lighting
engineering to describe source color attributes, and their effects on circadian wellness.
1.4.5.2 HUMAN SPECTRAL SENSITIVITY AND THE ENVIRONMENT
In humans and other organisms, light intensity and light rich in lower (blue) wavelengths are perceived by intrinsically
photosensitive retinal ganglion cells in the eye which signal light changes to the brain for suppressing the production and
secretion of melatonin (Bellia, Pedace, and Barbato 2014). The “sleeping hormone” melatonin is produced by the pineal
gland and is released during the night (Aubé, Roby, and Kocifaj 2013). A further discussion of the importance of the study of
circadian wellness as affected by the lit environment can be found in Section 2.3.1. Studies have used CCT to indicate many
blue-rich light sources which are not only capable of suppressing human circadian health, but also prone to exacerbated
scattering in the atmosphere adding to sky glow. Light pollution researchers have discussed the trend in outdoor lighting
practices to include more blue light sources (Lockwood, Floyd, and Thompson 1990).
A general progression from high pressure sodium to blue-rich LED lighting has been noted in the specification of outdoor
artificial lighting (Aubé, Roby, and Kocifaj 2013). Short wavelength light emitting from many outdoor lighting sources scatters
more than long wavelength light in some atmospheric conditions (Boyce 2014). This creates a potentially dangerous scenario
in dense urban areas where commercial and residential buildings achieve close adjacencies. In this scenario, networks of
outdoor fixtures provide an inescapable source of cool light trespass. According to artificial night light measurements taken in
Sherbrooke, Canada, illuminance measurements of 2 lux were commonly found in urban bedrooms. This value is slightly over
the 1.5 lux threshold observed to disrupt circadian rhythms (Aubé, Roby, and Kocifaj 2013). These findings, along with growing
trends in high-output, blue-rich LED outdoor lighting, suggest lighting-related health concerns may be on the rise. The design
Figure 1.3: Spectral Power Distribution Curves for Two Sources; Cool White Fluorescent (Left) and
Incandescent (Right) Light Sources (lrc.rpi.edu)
Chow | 21

of façade shading systems and lighting strategies may address the ability to eliminate light trespass at the building enclosure
plane.
1.4.6 ENVIRONMENTAL IMPACTS OF LIGHT POLLUTION
Researchers in the biological sciences have explored the relationship between the presence of artificial light in the
atmosphere and various types of flora and fauna. In the urban environment, the negative effects of this relationship are
compounded by increasing forms of light pollution. It is necessary to note that plant species react to a much broader range of
spectral content than humans. Light level measurements, though appropriate to discussing human responses to illumination,
may be inappropriate for use by ecologists. In his contribution to Ecological Consequences of Artificial Night Lighting,
Winslow R. Briggs discusses various types of plant photoreceptors, their functions, and the threats to germination,
phototropism, flowering, and dormancy. Amidst limited supporting research, experiments by researchers Cathey and
Campbell introduced lighting from conventional interior sources to plants at low light levels. The sensitivity to unnatural
light exposure varied by species, which, in the worst cases, affected plant’s ability to flower and prepare for winter dormancy
(Rich, Longcore, 2006). The influences of excess lighting on animal life have been extensively studied. The varying effects
of light pollution on animals can range from bothersome interference to fatal attraction. Whether near the ground or in the air,
light pollution sources present a danger to nocturnal mammals and birds which have developed sometimes deadly
relationships with the built environment.
1.4.6.1 NOCTURNAL MAMMALS
In regards to night lighting effects on nocturnal mammals, researchers have much more literature to reference. Lighting, or
more appropriately, its deficiency, has much to do with the patterns of daily activity, visual ability and physiological cycles
of these organisms. A number of disruptions may be caused during nocturnal mammals’ primary hours of daily activity (at
dawn and dusk), as well as longer-term effects to monthly and annual cycles (Rich, Longcore, 2006). Particular attention
must be paid to the effects of irresponsible lighting on the physiological adaptations of nocturnal mammals. The high
concentration of rods, photosensitive cells responsible for vision in low light, is one evolutionary advantage that is affected
by bright light. Above 120 cd/m², rods become saturated and unresponsive, causing temporary blinding effects (Rich,
Longcore, 2006).
1.4.6.2 BIRDS
The captivating effect of excessive lighting creates specific dangers for various bird species. The extensive study of bird
strikes, or bird-window collisions (BWC), concerns the attraction of birds to various properties of building facades, which
often results in injury or fatality. While the causes and frequency of bird strikes are vastly complex, a dangerous relationship
has been established between tall commercial buildings and birds migrating over urban areas. Spatial disorientation is
thought to occur when flying into bright lights at night causes a bird to lose its visual cues to the horizon. Using these cues to
inform their travel, birds make alterations to flight paths which “trap” them in the lighted area. In the presence of large,
sometimes light-emitting glazed facades, this effect results one of the most prolific dangers for birds in urban areas. Birds
trapped in the effects of façade lighting can collide fatally with structures, or else become victims of predation (Parkins,
Elbin, and Barnes 2015).
Conservative estimates number fatalities from bird-window collision in the United States between 365 and 988 million
annually, according to a recent analysis of published literature on BWCs and numerous unpublished datasets (Loss et al.
2014). Per building analysis of BWCs has been commonly conducted by researchers with minimal synthesis across studies.
The cross-resource analysis of 23 studies and over 92,000 bird fatality records conducted by Loss et al. accounts for year-
round assessment of bird strikes with residential buildings and developed estimates for low-rise and high-rise buildings. The
inclusion of building class aided in the distribution of data and the analysis of potential causes for bird strike. The following
Figure 1.4 categorizes the accumulated data by building type, according to this study.
Chow | 22

A recent study quantifying bird mortality in Canada estimates an annual total of the 25 million total fatal bird strikes (add
citation). Initial estimates for BWCs in Canada are disproportionally similar to that of the United States with regards to
building type. The study estimates 90% occur in the vicinity of single family residences, mobile homes, or duplexes. Nearly
10% of bird strikes occur in low-rise commercial and institutional buildings under 12 stories, while less than 1% occur
around tall buildings. Significant studies relating birdstrike to façade conditions are presented in Section 2.3.2.
The various research conducted to track the effect of façade lighting on BWCs has attracted the attention of lighting
designers. A recent design guide published by the Illumination Engineering Society of North America references Audubon
Society studies which report the significant reduction of BWCs in communities which switch off decorative building lighting
during migration seasons (Illumination Engineering Society of North America 2011).
1.5 STUDY BOUNDARIES
The absence of data on the subject of building contributions to light pollution formed the framework for the experiments
described in the following sections. Additionally, data collected in a neutral context of architecture may have the ability to
link light pollution sources and the negative impact on the environment. The study addressed the impact of a building’s
interior lighting on the environment, and the ability for façade design to limit this impact. Modeling the source of light
pollution as a building may help regulatory agencies shift necessary efforts to the reduction of light pollution.
This study addressed lighting and light loss, specifically light levels at various distances from a building façade. The data
produced were intended to be used by designers, architects, lighting designers, ecologists, sociologists, utilities, and
government agencies. The data can be referred to when addressing issues of light pollution and its effects on nearby
ecologies, residents, and communities. Due to the multi-disciplinary nature of this topic, the study carefully defined the role
of buildings as contributors to sky glow, glare, and light trespass in architectural terms. This analyses were accomplished by
physical and digital data collection methods for a glazed commercial building façade located in Los Angeles’s Downtown
area.

Figure 1.4: Estimates of Annual Bird Mortality by Collisions with U.S. Buildings (Loss et al., 2014)
Chow | 23

2 RESEARCH REVIEW: Building Contributions to Light Pollution
There are several important topics to consider when investigating the issue of light pollution and the effects of lighting at the
façade plane. First, it is important to understand how light pollution is understood and quantified, Secondly, it is necessary to
understand how artificial lighting acts upon a facade and third, how researchers are using building-sourced light pollution
data to apply to their research. The successes and oversights from the following research were essential to crafting a method
which accurately documented light pollution data in terms applicable to further study
2.1 LIGHT POLLUTION PERSPECTIVES: METHODS AND MODELING
Widening acceptance of excessive night sky brightness as a pollutant has inspired researchers in various fields to investigate
the potentially devastating consequences of over lighting in the built environment. The breadth of research which has resulted
from this attention has revealed the wide spectrum of understanding by which light pollution can be studied. A critical light
pollution study conducted by Luginbuhl et al. best described this spectrum in terms of “Sky-Down and Ground-Up”
approaches, illustrated in Figure 2.1 (Luginbuhl, Lockwood, et al. 2009). The researchers’ definition provided a framework
on which light pollution methods were evaluated and considered as relevant to study at the building scale.
2.1.1 SKY-DOWN APPROACH
The significant study conducted by Luginbuhl et al. in Flagstaff, Arizona estimated outdoor lighting on an urban scale,
utilizing land use data to determine per-unit lighting intensities by building type (Luginbuhl, Lockwood, et al. 2009). This
critical study describes the context in which similar urban studies have soured light pollution data. The “Sky-Down”
approach began as a purely observational method for understanding sky brightness. This approach utilized ground
observations and other data sourced from the night sky dome as an indicator of phenomena occurring many miles below.
Studies such as those pioneered by Treanor and Garstang often paired observations with a complex understanding of the
behavior of light encountering atmospheric contents to develop general sky models (Luginbuhl, Lockwood, et al. 2009).
The development of these sky models could only result in marginal information regarding the exact sources of artificial light
propagation. Garstang’s study of sky glow over the Los Angeles Basin and its effect on the Mount Whitney observatory
produced only estimates of light pollution sources in the urban environment. Similar studies resulted in indicators for
establishing Light Pollution levels which are still relevant astronomical light pollution metrics (Garstang 2004). Continuing
research in Italy has developed the Sky-Down approach, using night time satellite imaging to monitor and quantify light
pollution over urban areas. A recent study reviewed useful methods for monitoring and quantifying light pollution (Cinzano
et al., 2013). This study constituted a preliminary step in the quantification of light pollution beyond bare observational
evaluation of night sky brightness (Cinzano et al, 2013). Citing the dangerous potential for light pollution to affect the
environment and human health, Cinzano and Falchi proposed new methods for quantifying light pollution.
Figure 2.1: Flowchart Linking Sky Glow to Artificial Light Use
(Luginbuhl et al.,2009)
Chow | 24

The use of the Garstang modelling technique resulted in the generation of traditional indicators, including: upward flux, night
sky brightness at sea level, total night sky brightness, stellar visibility, loss of stellar visibility, and population-level statistical
indicators. These astronomical indicators were not fully appropriate metrics for the rapidly developing study of light pollution,
which has grown to encompass the environmental and health fields. Utilizing the Extended Garstang Model (EGM), Cinzano
and Falchi generated additional light pollution indicators as well as more detailed computation of classic indicators. The
researchers contributed a threshold for alteration, considering the atmosphere polluted when artificial night sky brightness or
radiation density is greater than 10% of the natural condition. They noted sky brightness alteration within 10% does not
necessarily mean the sky can be considered unpolluted. In these cases, additional indicators such as maximum luminance,
average hemispherical luminance, and average and maximum vertical illuminance are important to determine pollution levels.
Several Bi-Dimensional Indicators are clarified by the researchers:
Bidimensional Indicators
Often take the form of maps across the territory of the quantity in a given direction of sky, maps of the quantity
across the sky in an individual site, i.e. polar, Cartesian, hypermaps, etc. Several of these indicators can be easily
measured with satellite or astronomical data, and irradiance meters. The “single number” result of “integrated and
maximum indicators” are useful for quantifying light pollution at a site (Cinzano and Falchi 2014)
Upward light flux
Man made light flux emitted into the atmosphere, sourced by nighttime lighting installations
Artificial night sky brightness, luminance
The integral of artificial light scattered along an observer’s line of sight. Commonly computed at sea level
or ground level, this indicator affects the perceived luminosity of the sky, visibility of the stars, perception
of the universe, and the darkness of the environment.
Total night sky brightness
Shows the quality of the night sky in the territory as perceived by an observer, including the natural
components of the night sky brightness, commonly measured at zenith, accounting for the elevation
Star visibility
“Shows the magnitude of the faintest visible star, i.e. the capability of the population to see stars. Generally
it is computed at zenith, accounting for (a) the extinction of star light in the atmosphere from the top of the
atmosphere to the observer and (b) the eye ability in detecting point sources against a light background (i.e.
the polluted sky). This quantity is unsuitable to evaluate the light pollution of the atmosphere because of
the confusing effects of elevation and extinction: having a similar limiting magnitude on a mountain and in
the open ocean meaning that the mountain' sky is so polluted that the stellar visibility there is comparable to
that from sea level.” Variables that affect star visibility data include personal acuity, pupil size, confidence
of the observer, length of the observation, fatigue, and altitude.
Loss of star visibility
“The loss of star visibility (loss of limiting magnitude) shows the loss of the capability of the population to
see the stars, due to an increase in background luminosity.” The loss of star visibility is usually surveyed by
observers with “average experience and capability, aged 40 years, with eyes adapted to the dark, and
observing with both eyes toward the zenith”(Cinzano and Falchi 2014).
Number of visible stars
Generated as a new indicator by Cinzano because it is easier to understand than Star visibility. The data is
collected from “observers of average visual astronomy experience and capability, aged 40 years, with eyes
fully adapted to the dark, observing with both eyes the upward hemisphere and counting all the surely seen
stars”. Lost star maps are compared with maps of the same evaluated stars assuming no light pollution.
While useful for displaying the effects of light pollution, the researchers note these maps do not accurately
evaluate the ability to see the stars.
Sky irradiance
An Indicator which describes not only the irradiance on the ground surface, but also the irradiance of the
night environment as perceived by natural subjects, including plants and humans. For humans, this is
defined as illumination, which is weighted by the photopic visual response curve.
A significant aspect of the researchers’ work involved the development of 3-Dimensional monitoring techniques. An analysis
of a vertical section of the atmosphere above the Veneto plane was conducted along with the upward and downward artificial
radiation density at several horizontal sections at various altitudes of a test region. Taken above a polluted area of the Veneto
region, the horizontal sections pictured in Figure 2.2 document average density inside a unit of atmosphere at 47.5 km, 20.5
Chow | 25

km, 10.5 km, 4.5 km, 2.5 km, 1.5 km, 0.5 km. These images captured the nature of light scattering within the test volume of
atmosphere, and provided valuable insight into the distribution of light at various altitudes.

Figure 2.2: Upward (Left) and Downward (right) artificial radiation density Tb km
3
(Cinzano et al, 2013)
Chow | 26

Three-Dimensional Indicators
Often presented as a 3D array. These indicators can be obtained by detailed atmospheric modeling or direct
measurements via radiance meter, conducted with a balloon, airplane, or drone (Cinzano and Falchi 2014).
Radiation Density
The radiation density in the atmosphere is the number of photons in unit time (or the energy, or the light)
per atmosphere's unit volume in the course of transit, in the neighborhood of the points (x, y, z). This is the
best indicator for atmospheric light pollution because it can quantify how much content of natural light of
the atmosphere is altered by the introduction of artificial light. Radiation density is also useful in predicting
and quantifying the effects of urban lighting in the atmosphere's photochemistry (Cinzano and Falchi
2014). Radiation density may be described as radiation energy density (J m
-3
) Photon density (ph m
-3
), or
luminous energy per unit volume (Tb m
-3
) in CIE photopic and scotopic bands where Tb = lumens ×
second.
Upward and downward radiation densities
Useful indicators to describe the light directed back to the surface of the Earth due to its curvature, and the
light directed to space.
Radiation density due to direct illumination
Quantifies direct light from polluting artificial light sources travelling through a unit volume of atmosphere.
Upward and downward scattered flux
Describe the number of photons per unit time per unit volume (density of flux) rather than number of
photons per unit volume (density of radiation) in units ph s
-1
m
-3
Downward flux quantifies each unit volume of atmosphere as a secondary source of light pollution. The
density of downward flux can be expressed by the following equation, where F is the light flux scattered in
the direction ( θ; ϕ) by a unit volume of atmosphere in (x, y, z), obtained from LPTRAN.

Equation 2.1: Formula Used to Determine Downward Flux (Cinzano, Falchi 2001)
Chow | 27

Despite the development of “Sky-Down” light pollution study, satellite-based light pollution monitoring is not without its
limitations. The doctoral thesis developed by Alejandro Sanchez de Miguel investigated spatial variations in the techniques
for monitoring light pollution and its sources. Sanchez de Miguel discussed the value of research conducted by Cinzano et al.
which sourced satellite data from OLS cameras mounted to United States DMSP defense satellites as sources for calculations
of night sky brightness (Sanchez de Miguel 2015). Regardless of the highly advanced multispectral daytime (visible) and
nighttime (infrared) capabilities of the SUOMI/VIIRS camera which has succeeded OLS, limitations in remote sensing
techniques persist for light pollution quantification. Four VIIRS images illustrated in Figure 2.3 depict radiance variations
over Madrid Taken April 2012, October 2012, January 2013 and May 2014. These variations are likely due to instrument
errors which may be accounted for and corrected with continuing research (Sanchez de Miguel 2015).
Instrument field of view is another important concern for satellite-sourced data. The VIIRS view angle of 120 degrees
displays some test areas at angles up to 60 degrees from the vertical. Figure 2.4 illustrates this blocking concept in
pronounced viewing angles. It is important to note the obstruction by buildings for viewing angles far from nadir. Sanchez de
Miguel discussed analyses in which wide streets (7m) may be obscured by buildings 18m in height or taller. In such cases,
the relevant data range is reduced and areas outside an accurate field of view removed (Sanchez de Miguel 2015).
Sky-Down methods have pervaded many initial studies into light pollution, and have developed immensely with more accurate
equipment and dedicated techniques. Astronomical and astrophysical research have made use of this method to estimate light
pollution on a grand scale, but rarely include sensitive enough information to generate explicit understanding of source -specific
Figure 2.3: VIIRS/DNB Images over Madrid for April 2012, October 2012, January 2013 and May 2014(Right to left, Top to Bottom); Color
Scale Indicates Radiance in nW/cm
2
/sr (Sanchez de Miguel, 2015)
Figure 2.4: VIIRS/DNB Observations for Pronounced Viewing Angles May Be Obscured by Buildings Taller Than 6 Stories (18m)
(Sanchez de Miguel, 2015)
Chow | 28

contributions to light pollution. Data sourced from near-ground methods, by contrast, have demonstrated the potential to
regulate sources light propagation.
2.1.2 GROUND-UP APPROACH
At the other end of the light pollution quantification spectrum, “Ground-Up” processes have considered light output
measurements and work upward to estimate resultant forms of light pollution (Luginbuhl, Lockwood, et al. 2009). These
methods have made use of uplight percentages, wattage consumption, and partial luminous flux values to serve researchers
interested in wasted light or localized light losses. However, these methods have often been limited in their scope, and
difficult to relate estimates of total luminous fluxes. Luginbuhl et al. identified an exception to these studies which combined
street lighting and approximations of localized property light losses to estimate per capita lumen output.
Lockwood’s study at the Lowell Observatory took advantage of a wealth of recorded terrestrial factors to understand known
light pollution trends. Lighting regulations limited the lit environment to several sources, and provided a strong base from
which to estimate light output on the ground. Strict ordinances passed in 1989 Flagstaff, Arizona imposed lumens-per-acre
restrictions which contributed a similar base case for the team’s lighting assumptions (Lockwood, Floyd, and Thompson
1990). Using assumed luminaire quantities and initial lumen output, the team produced total lumens per capita which could
be attributed over a specified area. Horizontal illuminances could then be calculated to determine light levels arriving on the
ground plane. These figures received many more considerations. Natural night lighting factors were discussed, including
variations in light at the horizon due to the time of day, season, and meteorological conditions. This study utilized a strong
“Ground-up” method at the initial phase to develop simple, yet conceivable estimates of sky glow. Paired with available light
pollution data sourced by “Sky-down” approaches, Lockwood’s study expanded the potential influence of comprehensive
light pollution quantification methods.
2.1.3 COMPREHENSIVE STUDIES
While each of the discussed research questions operated within a designated scale and applicable topic, comprehensive
studies which approach light propagation issues from both ends of the quantification spectrum have often generated stronger
ties between light pollution effects and its causes. Studies which include Flagstaff, Arizona as a testing site consistently
approach light pollution from both ends of the spectrum, due to the city’s history of dark sky support and extensive light
pollution data. Luginbuhl’s 2009 study formed the first of a two-part study quantifying light pollution sources from light
reflected off the ground surface. A comprehensive land use survey, analysis of outdoor lighting sources, and list of post-
lighting code developments, and other “Ground-Up” sources were utilized to propose total sky glow figures. Compared with
pre-code estimates, the team developed predictive and speculative values for the area above Flagstaff (Luginbuhl, Lockwood,
et al. 2009). The second part of this study, published separately, incorporated a “Sky-Down” approach involving the team’s
initial sky glow measurements and a program implementing the Garstang model (Luginbuhl, Duriscoe, et al. 2009). The
comparison revealed a discrepancy between the estimates due to several critical assumptions held by the Garstang model.
These assumptions, which have much to do with the character of near-ground propagation phenomena suggested critical
information was needed to understand the role of the built environment in total sky glow. The Garstang model light level
estimates were more than twice the compared values. The team’s findings, based on analysis of Garstang’s model
assumptions identified near ground blocking by buildings and vegetation. It is important to note “considerable” spill light
originating from interior light sources was disregarded from the initial Part 1 study estimates (Luginbuhl, Lockwood, et al.
2009). Reasoning the small presence of commercial areas leading to this type of light leak through large windows, the team
considered their contributions in Flagstaff as negligible. Interior night light losses, however, have been proven to be
substantial in urban areas. Additional studies of building interaction in light propagation have continued to elaborate upon the
highly complex scenario occurring at the scale of the building facade.
2.2 LIGHT PROPAGATION: ILLUMINATION AT THE FAÇADE PLANE
The study of light propagation at the façade plane represents the shift from thinking along either end of the light propagation
spectrum and looking specifically at light pollution sources at the building scale. Special attention was paid to the
development of light pollution indicators at this scale and the definition of metrics by researchers in non-architectural
disciplines. While relatively light in terms of source material, there are several critical studies which investigate windows and
contributors to light pollution.
A critical study conducted in Tokyo, Japan studied the influence of interior lighting from office windows on direct sky glow.
This investigation of light pollution at the building scale involves the comparison of upward luminous flux from office windows
with the upward light sourced from outdoor lighting identified by the Illuminating Engineering Institute of Japan. The team
devised a method including both calculations and field measurements by which to collect building data (Oba et al. 2005). Spill
Chow | 29

light illuminance values were captured from ground level at a nearby park, which included light from office windows, outdoor
lighting, and existing sky brightness, while source illuminance values from solely the office windows were collected using
luminance values captured by the CCD Camera System as pictured in Figure 2.5.
The results of these calculations verified the measurements taken by the illuminance meter. Figure 2.6 illustrates the team’s
findings for changing illuminance values at the two park points over time for the various light sources. The team identified a
strong correlation between calculated and collected illuminance values. Two sets of calculations were performed with respect
to two distinct assumptions regarding the behavior of spill light. The first method, assuming upward and downward light as
equal, resulted in calculated values slightly lower than measured values. The second method, assuming downward lighting
consists of only direct lighting from the ceiling and its fixtures, resulted in calculated values slightly higher than measured
values. The research adopted the first method, adjusting the ratio of downward luminous flux to total luminous flux from 0.5
to 0.66, thus achieving results which equaled measured values. A rough check was conducted using a CCD Camera and
standard lens from two vantage points 24m above and below window level 200m from a building façade (Oba et al. 2005).
The experiment revealed the average ratio of downward luminous flux to total luminous flux to be 0.7, from a range of 0.69
to 0.73. The research team successfully devised a method of comparison between total illumination and illumination from the
building source based on physical measurements. These measurements were taken in a nearby park, where clear sources of
illumination could be identified and isolated. The researchers maintained reference measurements taken by illuminance
meter. The research also outlined the office buildings involved, generating the simple survey illustrated in Figure 2.7 of
building features to be used during the experiments. General statements on occupancy can be inferred from multiple
measurements taken to identify the effect of time on the percentage of illuminated windows.
Figure 2.5: Celestial Hemisphere at Point P as Captured by CCD Camera with Fisheye Lens Showing Lighting Sources From Outdoor
Lighting and Building Windows (Oba et al. 2005)
Figure 2.6: Changing Illuminance Levels from Outdoor Lighting Sources over Time for Two
Sources (Oba et al. 2005)
Chow | 30

The research team successfully devised a method of comparison between total illumination and illumination from the
building source based on physical measurements. These measurements were taken in a nearby park, where clear sources of
illumination could be identified and isolated. The researchers maintained reference measurements taken by illuminance
meter. The research also outlined the office buildings involved, generating the simple survey illustrated in Figure 2.7 of
building features to be used during the experiments. General statements on occupancy can be inferred from multiple
measurements taken to identify the effect of time on the percentage of illuminated windows.
The experiments conducted in the nearby park, however, lacked, “clean” data regarding light sources and their effects on the
collected measurements. Existing sky brightness was included in the values generated by illuminance meter, which cannot be
decisively purged from relevant illuminances. While the digital camera images illustrated light sources, there is little ability
to completely isolate per-building light loss. The inability to discriminate between the source buildings discounts any
variations in window construction and therefore, variations in light output. The successful measurements are indicators of
light pollution that acts most like light trespass because they are sourced from the ground.
A critical study performed by Kránicz was conducted to measure window spill light. The study was motivated by a large-
scale experiment which switched off street lighting and revealed the remaining building-sourced light to be a potential
contributor to light pollution (Kránicz, Kolláth, and Gyutai 2008). In the study, Kránicz identified urban features in
Veszprém in West-Hungary which were thought to contribute to light pollution and defined simple methods for photometric
analysis of these sources. The method for obtaining luminance data from pixels of images sourced from a professional
luminance imaging device suggested the importance of critical vantage points and distances from the measured surface. The
measurable quantities suggested by Kránicz and the research team provided a robust understanding of urban light pollution
objects including windows and dedicated façade lighting installations (Kránicz, Kolláth, and Gyutai 2008). Though several
dozen photos were taken, the research team included only four, prioritizing luminance values sourced at the “horizon.” In this
way, they might have overlooked potentially varied luminance data for other vantage points above and beyond the “middle”
of the window. In actuality, these reference points can be significant.
A low vantage point, for example, can include high luminances in the visual field, as an observer’s line of sight may include
a direct view of an installed interior luminaire. A vantage point above ground level, by contrast, can result in a line of sight
oriented toward floor surfaces that receive and reflect light from interior luminaires. A study of artificial lighting and its
interaction with windows was conducted by Darula et al., and clearly draws attention to secondary lighting sources as
potential sources of light pollution. His findings are summarized by Figure 2.8, which captures the general light patterns of
typical window constructions. It is important to note the division of the upper half-space and bottom half-space by the
Figure 2.7: Sample Building Survey
Chow | 31

centerline of the window shown below. Darula’s findings indicate variations in luminance measurements sourced from each
side of the window centerline (Darula 2013).
Light losses through windows were categorized in two parts: direct light which is generated by a luminaire and passes
directly through the façade and indirect light which is the product of multiple reflections from interior surfaces. Darula
addresses the behavior of interior light propagation based on the location of interior luminaires. The arrival of light upon the
inner window pane can be modeled by luminous flux Φ, as the result of the direct ( 𝛷 𝑑 𝑖 𝑟 ) and diffuse luminous flux ( 𝛷 𝑑 𝑖 𝑓 ).
𝛷 = 𝛷 𝑑 𝑖 𝑟 + 𝛷 𝑑 𝑖 𝑓
The results of Darula’s experiment regarding night luminance of a residential spaces’ window proposed the reduction
possibilities of conventional windows with and without curtains, and louvered windows, as observed during a clear night in
Bratislava from a distance of 54 m.

Figure 2.8: Light Propagation Patterns through Windows (Darula et al., 2013)
Chow | 32

Darula clarified the preliminary assumptions required for investigating the complex situation of light emission though glazing
and its shading conditions. Darula identified several variables for the behavior of light emissions, including: the quantity and
position of lights in the building, transmissivity of the glazing, and the structure and construction of any shading device. The
following diagrams depicted in Figure 2.9 illustrate the simplified effects of shading devices on internal light. Darula
acknowledged the difficulty of mapping façade area luminance and advised the generation of average façade luminance and
luminous intensity curves under ideal conditions.
Darula’s experiment measured light loss through residential windows, which may have significantly smaller glazing areas,
and may lack the complexity of glazing constructions used in commercial applications. The residential building type also
implied a certain level of occupation irregularity. The variable schedules of inhabitants created unpredictable differences in
measured window luminances over time. Despite these limitations, this experiment suggested nearly limitless exploration of
light loss through windows, particularly in terms of the reductive potential of associated shading technologies.
The need for light leak reduction techniques has become more relevant with the increasing concerns over light pollution by
professionals in various fields of study. Significant light pollution contributions from windows, which were noted by Luginbuhl
et al., and quantified by Kránicz and Darula, have been considered to varying degrees by researchers investigating its role in
medicine and ecology.
2.3 BUILDING SOURCED LIGHT POLLUTION: RESEARCH INTERESTS
2.3.1 CIRCADIAN WELLNESS AND THE BUILT ENVIRONMENT
A recent publication in the American Society of Heating and Air-Conditioning Engineers’ (ASHRAE) Journal described the
growing importance of light pollution research in the realm of the built environment. The primary motivation of this article
and many relevant studies concerns the significance of the circadian clock, an evolutionary advantage developed in many
organisms which relates proper biological processes to perennial cycles of day and night (Panda and Marks 2015). Health-
related responses to excessive light vary between animal species, and have been closely tied to sensitivity to wavelengths of
light. The 1958 discovery of shifting circadian rhythms in Gonyaulax polyedra due to blue light by John Woodland Hastings
and Beatrice M. Sweeny has gained importance in the study of similar sensitivity in humans (Holzman 2010). The various
reactions of the human body to pervasive blue-light exposure in urban environs include headaches, fatigue, medically defined
stress, anxiety, and decreases in sexual function (Panda and Marks 2015). Experiments involving humans, however, are
commonly conducted in controlled laboratory conditions. These surroundings lack consistent recording methods within the
subjects’ natural environments. Adjacencies to ubiquitous and dynamic urban light pollution sources are difficult to recreate
in a controlled testing facility. For this reason, the researchers proposed the use of wearable NIST-calibrated light sensing
devices to acquire 24-hour data on shifting lighting patterns with respect to participants’ surroundings (Panda and Marks
2015).

Figure 2.9: Idealized Effects of Various Shading Devices on Interior Light Loss Trajectories (Darula et al. 2013)
Chow | 33

2.3.2 BIRD WINDOW COLLISIONS
While research into the effects of light pollution on humans continues steadily, many critical studies have looked to buildings
as the source of disorientation for migrating birds. As discussed briefly in Section 1.4.6.2, bright light from the built
environment attracts migrating birds with sometimes fatal results. While broadcast towers, smokestacks, and other non-
building constructions have been studied as noteworthy collision structures, attention has recently been paid to interior night
light exfiltration through windows.
Various bird strike studies attempt to identify attractive elements of lit building façades. Researchers Parkins, Elbin, and
Barnes (2015), interested in the effect of interior night lighting on the frequency of bird-building collisions in urban areas,
focused their study on facades adjacent to Bryant Park in New York City. Identifying this urban park as a possible stopover
habitat for migrating birds, researchers monitored four sites under various building facades for 45 observation days. They
recorded each located carcass or injured bird as a single collision. The team conducted a light analysis of the project facades
on a weeknight one hour after sunset during the monitoring period and counted the number of illuminated window pixels
using Photoshop CS6 and Image J image processing programs. These illuminated pixels were compared with the total
number of façade pixels to generate a percentage of illumination. The pixel analysis conducted as quantification of the light
pollution sourced by windows is illustrated in Figure 2.10. It is important to note the lack of metrics indicating light intensity
in the analysis of the building façade. The research method was interested in proportions of lit to unlit portions of the façade.
The team multiplied this number by the number of floors found in each building to determine a light index and averaged the
building indices for each site. The sites were ranked by these indices and number of bird collisions using Spearman’s Rank
Order Correlation. The indices ranged from .11 to 4.67, the highest of which, were found to host the highest number of
collisions. These buildings were noted as having expansive and highly reflective glazing. This supported a positive
correlation between the proportion of glass and light index (Parkins, Elbin, and Barnes 2015).
Because lit windows and the highly glazed facades between 17 and 50 stories may have independent effects on bird-window
collisions, the team looked to previous research to make sense of these factors. A nighttime study in Manhattan showed few
collisions occurring during dark hours, which helped support the role of glass area as a major consideration for bird-building
collisions. It must be noted that the team does not discount the role of artificial lighting in bird-building collisions, which has
been known to disorient birds and may make them more susceptible to collision with reflective glass building surfaces.
Despite their findings that the percentage of glass in a building façade has an equal or greater effect on bird collisions in an
urban area, there are several light related factors left unconsidered by the experiments. The team categorized windows as lit
or unlit, without regard for light level. This categorization operates when assuming luminous intensity has no effect on bird
window collisions and that interior night lighting is consistent across the entire window area. This assumption was easily
made when measurements of illuminated windows are taken from the ground: a perspective which is not shared by the avian
test subjects. Variations in light intensity among the different buildings around Bryant Park were left unconsidered, and
allowed the research team to make a case for the importance of glazing area. Preceding research, however, has long held the
importance of lighting intensity as an attractive force for birds flying over urban environments.
Figure 2.10: Pixel Analysis of Illuminated Window Counts from a Single Vantage Point (Parkins,
Elbin, and Barnes, 2015)
Chow | 34

A recent study providing a first estimate of bird fatality due to collision with windows referred to the “beacon effect,” by
which light emitted by structures caused confusion and attraction to migrating birds active at night (Machtans, Wedeles, and
Bayne 2013). The study described the methods conducted to understand national bird population decreases and identify the
relationships between building types and their likelihood to cause avian mortality (Machtans, Wedeles, and Bayne 2013). The
team investigated single-family houses and duplexes, low-rise commercial buildings, and high-rise commercial buildings as
building types. The commercial building types were differentiated by a height of 12 stories; Buildings under this threshold
were considered low-rise commercial buildings, while those above were considered high-rise commercial buildings. Based
on a comprehensive literature review, the team generated a list of buildings based on likelihood to cause bird mortality (Table
2.1). The list of building types was complied with respect to amounts of assumed vegetation adjacent to sites and, perhaps
more importantly, large window area. These site and building factors occur in the majority of bird collision studies and
provided a clear bias in the generation of the researchers’ list (Machtans, Wedeles, and Bayne 2013).

Table 2.1: Number and Description of Buildings in Each Broad Class Used in Simulation (Machtans, Wedeles, and Bayne 2013)
Chow | 35

The team’s results were grand total estimates of bird mortality (Table 2.2). It is important to note the total number of bird-
window collisions at tall buildings is dependent on non-scientific survey data performed over a three-year period by the Fatal
Light Awareness Program (FLAP).
Despite low numbers of fatalities estimated for tall buildings, this building type demonstrated the highest per-structure kill
rate. Acknowledging the fatal relationship between birds and buildings, the researchers introduced the concept of mitigation
at the building façade. Strategies such as glazing frit patterns, films, and awnings were proposed to reduce bird attraction to
the “beacon effect” caused by interior lighting (Machtans, Wedeles, and Bayne 2013).
Programs in some U.S. states have accelerated light pollution reduction by calling for simple dimming or eliminating light
from unoccupied interiors during migration seasons. The Lights Out program initiated in Chicago resulted in an 80%
reduction in bird collisions at the McCormick Building’s façade (Zink and Eckles 2010). A study proposed by Zink and
Eckles sought to investigate light pollution reduction strategies as part of “Project Birdsafe” in St. Paul and Minneapolis and
propose a monitoring program to record collision data for these Midwestern cities (Zink and Eckles 2010).
To establish a more cohesive understanding of Bird-Window Collisions on western flyways, a comprehensive analysis of the
California Academy of Sciences (CAS) was conducted. The investigators sought to understand the many aspects of bird
biology that increase the likelihood of collision, the building design aspects which promote bird attraction, and the mitigation
strategies which actively reduce bird strike. The study reported significant preceding building-related research which
estimated high a frequency of collisions at lower windows where daytime bird activity is high, and acknowledged the threat
of collision for birds which migrate during the night (Kahle, Flannery, and Dumbacher 2016). As a contrast to other bird-
collision exploration results, collision monitoring at the CAS revealed low levels of bird strikes after 21:00 hours and a
significant peak at midday. Despite the majority of birds striking the building during daylight hours, it was found that
migratory birds were more susceptible to collision with building windows based on the ratio of collisions to local abundance.
Considering the night-time migration patterns of these highly susceptible birds, night time light losses may be still be
considered a significant part of bird strike.
In addition to the thorough collection protocol implemented at the site, an extensive survey of the building facades was
conducted. This allowed windows to be analyzed with respect to their overall size. Small pane and large pane windows on the
south façade received disparate rates of bird strikes. Large panes of glass received nearly 10 times more fatal bird strikes per
unit glass than small pane windows. The relatively equivalent glazed areas between small (1237 m
2
) and large (1143 m
2
)
windows allowed the window types to be easily compared. These findings confirm similar findings from Bryant Park by Parkins
et al., which indicate glazing area may be an important and possibly independent factor in bird disorientation.
The survey was also a critical aspect in the development of a mitigation strategy to be tested along the east and west
windows. Retractable vertical shades were utilized to block exterior glass areas above 3.5 m, or 2/3 the window area (Kahle,
Flannery, and Dumbacher 2016). The shade strategy covering 2/3 of the window are is illustrated in Figure 2.11.
Table 2.2: Summary of Estimates of Bird Mortality Caused by Bird-Window Collisions at Different Types of Buildings in Canada
(Machtans, Wedeles, and Bayne 2013)
Chow | 36

The results of intervention revealed pre-mitigation strike rates were 22 times higher than the shaded condition. This drastic
difference suggested an important, though non-linear relationship between reduced façade glazing area and the reduction of
bird strike. Similar façade mitigation procedures resulted in bird collision decline. The introduction of mullions at every 0.5
m for even large expanses of windows resulted in reduced bird strike (Kahle, Flannery, and Dumbacher 2016). The
researchers discussed the reduction of window pane size, the installation of exterior shades, and the application of false
mullions to be cost-effective design strategies for future and existing buildings. It must be noted that while light leak and
large expanses of glazing area may attract and disorient migrating birds in different ways to similarly disastrous ends, a
singular shading strategy may potentially limit light exfiltration and limit reflective glass area. The result of implementing
façade strategies to mitigate these two seemingly independent bird attractors may address two issues with a single solution.
2.4 RESEARCH SYNTHESIS
My review of the literature highlighted the importance of a multi-disciplinary approach to understanding and reporting the
effect of façade constructions on the propagation of urban light pollution. The synthesis of the successes of the investigated
research has influenced the design of the three phase methodology (Chapter 3).
As a complement to understanding light pollution measurements taken from satellite sources, Cinzano and Falchi (2014)
identified the importance of understanding light propagation through horizontal sections measuring artificial radiation
density. A dissected view of the elevational variations in light propagation would prove instrumental in understanding light
loss over a multi-level building façade. This suggested the possibility for depicting changes in light propagation patterns from
the ground floor to upper floors, possibly taking into account changes in building geometry or obstructions occurring along
the way. The researchers considered light propagation in terms of radiation density separated by kilometers, which may be
unrealistic parameters for a study whose main audience is building professionals rather than astronomers. Their clarified
definitions for non-traditional three-dimensional indicators inspired the spatial analysis of light pollution propagation.
The work of Kránicz and the research team (2008) defines the methods for measuring light pollution data in simpler, more
accurate terms than the top-down satellite approach. Conventional photometric analysis results in easily obtainable data
which can be understood by lighting designers and engineers familiar with luminance imaging and illuminance
measurements. The methodology employed by the researchers suggested the importance of critical vantage points from the
façade, and help inspire the development of the physical collection methods. The team identified a clear table of calculated
quantities which was imperative to the development of a survey and table of results which could be used by other disciplines.
The research team studying the effect of interior night lighting on exterior lighting successfully devised a method of
comparison between total illumination and illumination from the building source. These physical measurements, though not
based on material or geometric properties, introduced the formulaic approach to lighting calculations based on photographic
data. Various reduction factors specified in the calculations offered generic transmissivity factors for glazing and generalized
multipliers attributed to the differences between upward and downward lighting. These variations reinforced the importance
of digitally testing variations in transmittance which are inherently material conditions. The simple survey reinforced the
importance of understanding the lighting quality, strategies, and sources in the building interior. This was extrapolated to
include more detailed information regarding interior finishes and reflectance values. Variations in light output during dark
hours informed the possible peak times to be tested during the point-in-time physical testing process (Kránicz, Kolláth, and
Gyutai 2008).
Figure 2.11: Photos of the East Side Windows without Exterior Shades (Left) and with Exterior Shades (Right),
(Kahle, Flannery, and Dumbacher, 2016)
Chow | 37

Darula’s research clearly draws attention to the emission of artificial light into the atmosphere as a geometric process through
a window. Darula clarified the preliminary assumptions required for investigating the complex situation of light emission
though glazing. Summarized by clear figures, these simple diagrams influenced the design of the testing grid, which took into
account the upper, lower, and middle regions of the window (Darula 2013). A view normal to the window center was also
mandated by the research, which influenced both the use and placement of a digital camera in luminance measurements.
Darula’s analysis of light patterns of typical window constructions validated the need for digital simulations of various
geometric conditions (Darula 2013).
Bird Collision research conducted by Parkins, Elbin and Barnes in New York’s Bryant Park describes a need for accessible
and actionable light pollution data for use by invested professionals in other disciplines. Despite their findings that the
percentage of glass in a building façade has an equal or greater effect on bird collisions in an urban area, their analysis of the
building façade offered clues as to relevant information that may be provided by future light loss experiments. The research
clearly employed per-pixel analysis of illuminated windows, which could easily be replaced by more complex interior light
data imaging (Parkins, Elbin, and Barnes 2015). The accumulation of relevant data, as collected by Kránicz and Oba in their
respective endeavors could be useful tools for understanding interior night lighting and its effects on the frequency of bird-
building collisions (Oba et al. 2005).

Chow | 38

3 METHODOLOGY
To understand the loss of interior night lighting through a commercial building façade, a methodology was generated to
validate digital simulations with field data from an exemplar case study. The analysis of an existing case study was
imperative to understanding exfiltration and validating a digital model. The validated model was used to simulate real world
and speculative shading scenarios. The totality of the project consisted of three phases, summarized in Figure 3.1, following
selection of an appropriate case study building. The Data Collection phase included the survey analysis of the building case
study and the capture of illuminance and luminance data via physical experiment. In the Data Processing phase, Rhinoceros
5.0+DIVA 3D Modeling software was used to replicate the building case study in detail and simulate the existing exfiltration
phenomena. The results of the data processing stage confirmed ability of the DIVA Plugin to accurately simulate reality. The
third phase included digital modeling of various façade shades and the accurate simulation of their potential to reduce
unwanted light loss.

3.1 CASE STUDY SELECTION: 800 WILSHIRE BOULEVARD
The physical experimentation which formed the core of the Data Collection phase required the selection of an appropriate
building case study. The following selection criteria represent desirable parameters for studying the effects of light loss in an
urban environment as well as considerations for convenience and accessibility:
Building Case Study Selection Criteria
The goal of the selection criteria was to locate an appropriate building with which to test light loss contributions to
the exterior environment. The criteria were chosen to prioritize buildings that could have the most lighting impact,
and therefore, a high potential for mitigation in the urban area of Los Angeles. Even with a rich collection of
commercial building types and geometries in the downtown area of Los Angeles, the selection of a building case
study was not without its logistical considerations.

Figure 3.1: Three-Phase Methodology
Chow | 39

Multi-story Commercial Building
The program of the chosen building represents the visual trend of lights left on in downtown buildings
during night hours. Unlike residential towers in the same urban center, commercial towers have consistent
patterns of occupancy with most occupants leaving the building after business hours. Lights left on in
vacant office building spaces not equipped with occupant or time-based vacancy sensors are likely to stay
illuminated until the occupant returns the next work day. The exception is, of course, the decision of a
custodial staff member whose schedule was not factored in point-in-time analysis of illuminated lights.
Curtain Wall Facade
The decision to study a particular building with a curtain wall façade and Window-to-Wall Ratio above
30% ensured the material transparency of the chosen building would have greater potential to allow escape
of interior night light. The 30% cutoff did not exclude several of Los Angeles’s older buildings, in case
permissions could be attained for a site without a contemporary glass façade.
Buildability via Digital Modeling
In direct consideration of both modeling and simulation time, a case study building with a simple façade
geometry was preferred.
Nonadjacent to Outdoor Light Sources
To preserve the integrity of the light level measurements, the proposed building case study was intended to
be far from light emitting sources such as landscape and roadway lighting.
Access to Specification Data
The chosen building would have easily-attainable or previously published documentation including
architectural drawings, performance criteria, and interior lighting specification data to aid in the 3D
modeling process.
Access to Façade for Light Level Testing
The critical criterion for selecting the case study would be access to the façade for initial testing. This
includes permissions from building operators, facilities managers, architects, and/or tenants to test the
building façades adjacent to commercial spaces. In consideration of cost and liability, no scaffolding or
special equipment should be used to access the façade from an upper floor of the building. For this reason,
Priority selection would be reserved for those buildings whose facades were accessible from ground level,
and whose ground-floor façade construction most closely matched the construction on the upper floors. The
interior spatial and lighting environments would be easily repeatable from floor to floor.
After discussions with building administrators and tenants of various buildings on the University of Southern California
campus and the neighboring downtown Los Angeles area, the 800 Wilshire Building in Downtown Los Angeles was selected
as a building case study.
Due to the generous cooperation of researchers at Buro Happold Engineering, the location of the company’s 16
th
floor offices
at 800 Wilshire became a near-ideal case study for analysis. In addition to meeting the initial selection criteria, the
collaboration provided access to building plans and dimensions, office material information, and accurate lighting
specification data. The terraces on the building’s top (16
th
) floor permitted full access to the façade from the exterior. An
initial decision to test the building façade at the ground level considered the dangers of mounting scaffolding and the cost of
using drone or balloon to assist in the data capture at higher elevation. The opportunity to test at an elevation high above
street level invalidated this initial consideration. Measurements taken at this appropriate commercial building height
eliminated the allowances which would have been made for variations in lighting fixtures, spatial dimensions, and façade
constructions or materials between the ground and upper floors. The building site is pictured shown in Figure 3.2.

Chow | 40

Figure 3.2: Location of Case study Building at 800 Wilshire Blvd, Los Angeles, Google Maps 2015
Located on the corner of Flower Street and Wilshire Boulevard, the 16-story commercial tower is attributed to noted Los
Angeles architect Welton Beckett (Vincent 2004). Construction on the 226,797 square-foot tower completed in 1972 and was
recently renovated after acquisition in 2013 by a developer and investment firm. The multi-year renovation focused on the
“major transition” of the downtown market, updating the main lobby and building finishes to attract creative tenants to the
downtown area. Notable enhancements to the 16
th
floor offices include exposing the overhead systems in an open-ceiling
model, new lighting, and finishes (“Lincoln Property Completes Renovation of 800 Wilshire” 2015). Additionally, non-
conventional open-plan work spaces were organized adjacent to the south and north balconies pictured in Figure 3.3.

Figure 3.3: Buro Happold Offices on the 16
th
Floor of 800 Wilshire; Open Office (Left), and Balcony (Right)

Chow | 41

3.2 PHASE 1: DATA COLLECTION
Physical data collection was performed to accurately document the results of light loss through an existing façade in
downtown Los Angeles. Data collection began with pre-selection analysis for the selection of a case study and a detailed
survey of the existing interior conditions, design and construction of the data collection module used in the physical
experiment, purchase and training of the necessary technology for light level documentation, and processing of the resultant
light loss data. Though much of the inspiration for the experiment design was gleaned from the combined literature
synthesized in Section 2.2.1, the resulting procedure for capturing light level measurements required tailoring to the selected
building case study. The diagrammatic summary of this Phase 1: Data Collection procedure is depicted in Figure 3.4

Figure 3.4: Diagrammatic Summary of Phase 1 Data Collection Procedure. Note the Survey and Photometric Analysis Required Minimal
Software Processing.

Chow | 42

3.2.1 MEASUREMENT
3.2.1.1 SURVEY
The intention of the survey was to assemble as much possible information about the building for subsequent 3D modeling.
The most important overall building considerations are listed in the following Table 3.1. Considerations for the interior
spaces’ dimensions, finishes, and lighting strategies are listed in Table 3.1 and Table 3.2.
Table 3.1: Important Case study Building Survey Considerations
Facade Features Building Features
Footprint Testing

Height m Height m Access Y/N
Width m Width m Distance from Façade m
Area m
²
Length m Height m
WWR %
Window Type -
Window Transmissivity %

Table 3.2: Case study Conditions Survey
Finishes Lighting
Luminaire Control Lamp

Floor - Length m Dimming Y/
N
Number #
Wall - Width m CCT #
Ceiling - Mounting Height m Lumen Output #
Type - IES file? Y/N
Existing Shading Manufacturer - Manufacturer -

Chow | 43

3.2.2 DATA COLLECTION MODULE
The design and construction of the Data Collection Module was an early concession to preserve the integrity of the on-site
light level measurements. The intention with this module was to block out all exterior light as well as limit the testing area to
a particular section of the facade. Theoretically, the construction could be expanded to accommodate larger façade areas. The
information gathered from the initial design survey was critical to developing the construction, whose critical dimensions are
depicted in Figure 3.5.

Figure 3.5: Data Collection Module Design and Critical Dimensions.
The survey data relating to the façade construction directly influenced the design of the collection module. For instance, the
width of the window bay, being one meter, was translated in the width of the collection module. Additionally, the depth of
the model was dictated by the dimensions of the case study building’s balcony. This allowed for two additional measurement
distances at 1 meter and 2 meters from the exterior façade plane.
The construction of the module was accomplished using ½” Polyvinyl Chloride (PVC) piping and various appropriate right-
angle fittings. The material was carefully measured to accommodate insertion into the fittings and carefully cut with a simple
hand saw. The PVC frame was then outfitted with 16 yards of medium and heavyweight cotton blackout fabric that was cut,
sewn, and slipped over the pipes making up the frame. These panels were hung to accommodate an overlap at the edges.
Clamps held together the panels during testing. The black box was shaded on 4 sides, leaving the base and plane closest to
the façade unshaded. Also marked in the above diagram are the locations of the testing nodes, which were placed equidistant
to each other within the extents of the façade area bounded by the box. For reference, the diagram in Figure 3.6 shows the
location of the Data Collection Module and its relevant dimensions.
Figure 3.6: Data Collection Module Erected in Place at the Case study Façade Plane. Note the
Locations of Equidistant Test Nodes within the Module.
Chow | 44

3.2.3 PHOTOMETRIC ANALYSIS
Photometric Analyses were conducted after dark during business hours on November 23
rd
2015 and December 11
th
, 2015.
Luminance photos and illuminance measurements were taken from the inside of the Data Collection Module assembled on
exterior 16
th
floor south terrace.
3.2.3.1 LUMINANCE PHOTOGRAPHS
Photos were taken at a vantage point 1m from the façade using a Nikon D300 with a fisheye lens. The camera’s automatic
bracketing feature was enabled before testing. A remote shutter button was used to minimize movement of the camera once
in place. The resultant photographs preserved narrow ranges of exposure in the default file extension for images taken with
Nikon’s D300: .NEF. After processing in Photoshop, the image files were saved in .HDR format. The resultant High
Dynamic Range file depicted in Section 4.1.6, was then converted to false color for analysis in hdrScope, a utility developed
by Viswanathan Kumaragurubaran in collaboration with Mehlika Inanici at the University of Washington, Seattle
(Kumaragurubaran 2012).
3.2.3.2 ILLUMINANCE MEASUREMENTS
Illuminance Measurements were taken with a NIST certified EXTECH 407026 lux/foot-candle light meter at predetermined
calculation points within the Data Collection Module. The calculation points were demarcated with a temporary grid of
monofilament line following the capture of luminance photos at vertical planes 1m and 2m from the façade plane. Several
additional illuminance measurements were taken at the interior and exterior façade glazing planes for use in digital model
validation. All illuminance measurements were recorded manually. The following Figure 3.7 is a composite image of the
Data Collection Module in place on the 16
th
floor terraces.

Figure 3.7: Composite Photograph of Data Collection Procedure
Chow | 45

3.3 PHASE 2: DATA PROCESSING
The digital modeling procedures in the data processing phase required advanced knowledge of Rhinoceros 5.0 and the
DIVA4Rhino simulation plugin. Information from the initial building survey was required at the modeling phase to generate
an accurate digital replica of the space. Simulations in DIVA were conducted to achieve illuminance measurements and
luminance renderings capable of being compared with photometric analyses. A diagrammatic summary of the simulation-
based Phase 2: Data Processing procedure is depicted in Figure 3.8.
3.3.1 MODELING
The Modeling process in Rhinoceros 5.0 required the careful translation of information from the initial building survey
conducted in Phase 1. Interior and exterior measurements were used to accurately build the digital model. Several spatial
obstructions which could affect simulation accuracy including structural columns, furniture, and a partial drop ceiling were
extruded as simple geometries. All simple surfaces were modeled with directional vectors oriented to the interior and divided
into layers based on material. Figure 3.9 illustrates the digital model replicating the case study building. It is important to
note the window test area and dedicated layers based on material.
Figure 3.8: Diagrammatic Summary of Phase 2: Data Processing Simulation Procedures Note: Relevant Finish and Lighting Strategy
Information from the Initial Building Survey
Figure 3.9: Screenshot of the Digital Model Replicating the 16
th
Floor Terrace.
Chow | 46

3.3.2 DIGITAL SIMULATION
Digital Simulation using DIVA4Rhino required a dedicated workflow. For the simulation model replicating the conditions at
800 Wilshire, the following parameters were addressed:
Location
A weather file in the conventional .epw file format was specified during the initial preparation of the simulation
processes. The selection of the site-specific weather file generated the DIVA directory folders in the same file
location as the Rhinoceros file.
Nodes
The generation of illuminance calculation nodes required the development of horizontal calculation planes. Three
1m-wide by 2m deep calculation planes were generated as source planes. The specified spacing of .425 for each
horizontal plane resulted in equidistant node locations corresponding to the physical measurement points, plus
intermediate points located at 0.5 m and 1.5 m from the façade plane. The node locations are illustrated in Figure
3.10.

Materials
The selection of material properties was imperative to simulating accurate results for the Data Processing Model.
The definitions of these materials were based on manufacturer supported models and values when available.
Assign Materials
Custom Radiance Glazing (GLASS) materials and translucent surface (TRANS) materials were generated
to ensure accurate simulation of the specified SolarGard film and Mechoshade retractable window shades
The material definitions, discussed at length in Section 5.2.1.2, were added to the model-specific materials
data file, located in the resources folder. Materials were specified by the layers set up in Rhinoceros.
Load IES Files
An .ies file for Finelite’s lensed HP4 was imported into the model, placed at the proper mounting height
and replicated to reflect the existing space conditions as specified in the tenant-provided reflected ceiling
plan.
Metrics
The simulation processes required to calculating illuminance values and luminance visualizations are detailed by the
following DIVA Commands:
Daylight Images
Found under the Daylight Image tab, the Visualization Simulation was the main operation used for
rendering visualizations, calculating luminance values, and generating false color renderings. Resulting
illuminance values were recorded in a spreadsheet for validation efforts.
Daylight Grid-Based
Figure 3.10: Calculation Node Locations for Data Processing Digital Model, Note the Intermediate Nodes and
Equidistant Point Spacing
Chow | 47

Found under the Daylight Grid-Based Simulation Categories, Point-in-Time Illuminance was the main
operation used for generating illuminance values on the calculation planes specified by the physical data
collection points. The resultant illuminance values were recorded in a spreadsheet format for validation
efforts.
3.3.2.1 VALIDATION OF THE DIGITAL MODEL
The validation of the digital model was accomplished by comparing the digital results with physical light data. The accuracy
of illuminance calculations was evaluated by comparing peak values at each calculation plane. The peak values were recorded
with respect to the source plane, and offered reliable data regarding the reduction and distribution of light through the glazing
material. Averaged illuminance values, though useful when comparing multiple reduction strategies, would convey an
incomplete understanding of light distribution. Luminance was evaluated by comparing peak luminance measurements labeled
by the DIVA4Rhino wxfalsecolor utility. Fidelity to the measured values was described as a percentage.
3.4 PHASE 3: REDUCTION TESTING
The Reduction Testing process required advanced knowledge of Rhinoceros 5.0 and the DIVA4Rhino simulation plugin as
well as a working knowledge of WINDOW 7.4 and OPTICS6. A “shoebox” model was developed to reflect a conventional
office space and lighting strategy. A baseline “unglazed” condition was simulated for comparison with various façade
shading materials substituting the glazing layer. These were then compared to one another and ranked in terms of their
reduction potential.
3.4.1 MODELING
The software packages were developed by the Lawrence Berkeley National Laboratory for the purpose of analyzing digital
window constructions. WINDOW 7.4 was used to generate optical data, including visible light Transmittance (VLT) values
for manufacture film and glass types. These values were translated to transmissivity values for use in generating custom
Radiance materials. OPTICS6 was used to generate complex Radiance definitions for complex glazing constructions.
The Modeling process for the baseline model in Rhinoceros 5.0 required the careful translation of information from reliable
building images. The space interior, model after a conventional single office space in the building, was designed 3.5m wide
and 3.5m deep, with a ceiling height of 3m. These dimensions corresponded to an office width spanning 4 façade glazing
units. All simple surfaces were modeled with directional vectors oriented to the interior and divided into layers based on
material. Figure 3.12 illustrates the digital model representing the environment for testing reduction potential.
3.4.2 DIGITAL SIMULATION
Digital Simulation using DIVA4Rhino required a dedicated workflow. For the simulation model replicating the conditions at
800 Wilshire, the following parameters were addressed:
Location
A weather file in the conventional .epw file format was specified during the preeminent step of the simulation
processes. The selection of the site-specific weather file generated the DIVA directory folders in the same file
location as the Rhinoceros file.
Figure 3.11: Diagrammatic Summary of Phase 2: Data Processing Simulation Procedures Note: Relevant Finish and Lighting Strategy
Information from the Initial Building Survey
Figure 3.12: Screenshot of the Digital Model, Note the Layers Generated by Material
Chow | 48

Nodes
The generation of illuminance calculation nodes required the development of horizontal calculation planes. five 1-m
wide by 2-m deep calculation planes were generated as source planes. The top and bottommost planes were placed
at the upper and lower limits of the façade unit, while three calculation planes were placed equidistant along the
glazed area height. The specified spacing of .220 for each horizontal plane resulted in equidistant node locations
corresponding to the physical measurement points and previous digital simulation. The node locations representing a
much denser and therefore, more cohesive calculation grid are illustrated in Figure 3.13.

Materials
The selection of material properties was imperative to simulating accurate results for the Data Processing Model.
The definitions of these materials were based on manufacturer supported models and values when available.
Assign Materials
Custom Radiance Glazing (GLASS) materials and translucent surface (TRANS) materials were generated
to ensure accurate simulation of the specified façade shading strategies. These include custom glazing types
for various 3M films. The material definitions, discussed at length in Section 5.2.1.2, were added to the
model-specific materials data file, located in the resources folder. Materials were specified by the layers set
up in Rhinoceros.
Load IES Files
An .ies file for Cooper’s 2ftx4ft 2GCAML Series Louvered troffer was imported into the model, mounted
at ceiling height. The file was replicated to reflect the conventional lighting strategy installed in the office’s
tile ceiling.
Metrics
The simulation processes required to calculating illuminance values and luminance visualizations are detailed by the
following DIVA Commands:
Daylight Images
Found under the Daylight Image tab, the Visualization Simulation was the main operation used for
rendering visualizations, calculating luminance values, and generating false color renderings. The
visualizations were taken by a “camera” placed 1m from the façade with view properties matching a 180º
fish-eye lens. Resulting luminance values were recorded in a spreadsheet for comparison.
Daylight Grid-Based
Found under the Daylight Grid-Based Simulation Category, Point-in-Time Illuminance was the main
operation used for generating illuminance values on the calculation planes specified by the physical data
collection points. The resultant illuminance values were recorded in a spreadsheet format for comparison.
3.4.2.1 COMPARISON OF REDUCTION POTENTIAL
Figure 3.13: Screenshot of the Digital Model Describing a Glazed Condition for Reduction Testing
Chow | 49

The comparison of the reduction potential of the proposed façade shading systems was accomplished by paralleling the peak
illuminance and luminance results. The reduction potential of each façade was described as a percentage relating the tested
shade to the unglazed condition.

Chow | 50

4 RESULTS
Chapter Four contains the results for the Physical Data Collection and the Digital Simulation in Phases 1 and 2, as well as the
Digital Simulations of Reduction Potential in Phase 3. The results of the survey provided important building information for
use in planning and designing the physical experiments in Phase 1. In addition to general building dimensions and space
planning, information regarding the lighting strategy and sources, installation of window films, and shading devices were
necessary to obtaining accurate results. Results from Phase 1 became an accurate snapshot of light exfiltration at the
commercial building scale. The collected values represented various possible conditions in a contemporary downtown work
environment. These values served as an initial benchmark to validate the accuracy of the digital model and simulation
completed in Phase 2. The High Dynamic Range (HDR) image results of the simulation generated with Rhinoceros
5.0+DIVA are included in Section 4.2.2. The results of the Phase 3 Reduction Potential testing of various façade
constructions provided measureable alternatives to unmitigated interior night lighting. The results of this reduction testing are
listed in Section 4.3. The following results have been separated by phase, following a short summary of pertinent variables
and a discussion regarding units and format.
4.1 PHASE 1: CASE STUDY DATA COLLECTION
4.1.1 SURVEY
The visual survey was conducted after selecting 800 Wilshire Blvd as the case study location. The diagrammatic results of
the visual survey depicted in Figure 4.1 represent information regarding the lighting strategy, material selection, façade
measurements, and shading.

The lighting strategy above the open-concept workspaces uses indirect lighting to provide general illumination. Any task
lighting needed within the spaces was achieved by lamps which were noted as unused during testing hours. They are included
in neither the lighting survey in Table 4.1, nor the digital simulation.

Figure 4.1: Diagrammatic Results for the Visual Survey
Chow | 51

Table 4.1: Physical Data Collection; Lighting Survey Results for 16th Floor Open Offices
Lighting Survey

Description Manufacturer Catalog Number Qty.
Fixture

HA1 4” Recessed High Performance
LED Luminaire
Finelite HP-2D-16-HO-4000K-277-
FMMC
8

Control
Pico
Remote
Wireless Dimming Switch Lutron PJ-2BRL-WG-I01- 8

The material finishes in the space provided a wide range of reflectance values. The limitations of the DIVA for Rhinoceros
Simulation plugin, as discussed in Chapter 3, were taken into account when deciding approximate reflectance values. The
results for the visual survey are summarized in Table 4.2.

Table 4.2: Physical Data Collection; Material Survey for 16th Floor Open Offices
Material Survey

Color Material Approximate Reflectance
Value
Wall White Painted Gypsum 80
Exposed Ceiling Black Painted Stucco 0
Soffit White Painted Gypsum 50
Floor Gray Carpet 20

The following Table 4.3 provides information regarding the window film and mechanized roller shades used on the south
facades of the open office. The locations of the mechanized window shades are shown in Table 4.1.

Table 4.3: Physical Data Collection; Shading Survey for 16
th
Floor Open Offices
Shading Survey

Description Manufacturer Transmissitance
(T%)
Solar Gard
Panorama Slate 20
Adhesive Solar Control Window
Film
Solar Gard 83

Mecho Shade
SunDialer System
Mechanized Rollershade, 5%
Open Basket Weave
MechoSystems 5

Chow | 52

Measurements taken from the façade, imperative for accurate modeling in the digital environment, are summarized in Figure
4.2, which also describes façade material considerations. Important considerations include interior obstructions such as
interior columns and spandrel panels.

Figure 4.2: Critical Façade Dimensions and Considerations Required for the Digital Modeling Process
Chow | 53

4.1.2 ILLUMINANCE
The illuminance measurements captured light levels as interior light exited the façade, landing upon the defined horizontal
and vertical calculation planes. The data outlined in Table 4.4 were collected on November 23
rd
, 2015 at 15:30 using a NIST
certified EXTECH 407026-NIST lux/foot-candle light meter with a resolution of 1 lux. This model’s built-in calibration
adjustment for various light sources does not require a correction multiplier. The metric used for the following measurements
is the SI unit lux.
4.1.3 SOUTH FAÇADE UNSHADED
Table 4.4: Physical Data Collection; Illuminance Measurements, South Facade Unshaded
Illuminance Measurements
(Lux)

So. Façade
Unshaded
1
A B C
a 4 1 0
b 4 1 0
c 4 1 0
2
a 4 2 0
b 4 2 0
c 4 2 0
3
a 4 0 0
b 4 0 0
c 4 0 0
0
a 0 0 0
b 0 0 0
c 0 0 0

Chow | 54

The illuminance measurements captured light levels as interior light exited the façade, landing upon the defined horizontal
and vertical calculation planes. The collection of these data points were achieved after the mechanized shade was lowered
using a dedicated wall switch on the interior. The data outlined in Table 4.5 were collected on November 23
rd
, 2015 at 15:30
using a NIST certified EXTECH 407026-NIST lux/foot-candle light meter. This model’s built-in calibration adjustment for
various light sources does not require a correction multiplier. The metric used for the following measurements is the SI unit
lux.
4.1.4 SOUTH FAÇADE SHADED
Table 4.5: Physical Data Collection; Illuminance Measurements, South Facade Shaded
Illuminance Measurements
(Lux)

So. Façade
Shaded
1
A B C
a 2 1 0
b 2 1 0
c 2 1 0
2
a 2 1 0
b 2 1 0
c 2 1 0
3
a 2 1 0
b 2 1 0
c 2 1 0
0
a 0 0 0
b 0 0 0
c 0 0 0

Chow | 55

4.1.5 LUMINANCE
The results of luminance analysis are bracketed images taken at various exposure values. They are shown grouped together in
the following figures by façade shading condition.
4.1.6 SOUTH FAÇADE UNSHADED
Figure 4.3 South Façade, Unshaded Bracketed Images taken with a Nikon D300
Figure 4.4: Physical Data Collection; So. Facade Unshaded HDR Composite Photo
Chow | 56

The hdrScope software tool translated these images into false color luminance maps with labels describing peak luminance
data in the units candela/m
2
. This false color Luminance map, depicts peak luminance along a scale determined by the
extreme values found in the image. A scale of 0–3000 candela/m
2
was specified for the map depicted in Figure 4.5. It is
important to note the peak luminance of the unshaded condition represents the maximum theoretical luminance value, as the
shaded condition effectively reduced luminance.

Figure 4.5: Physical Data Collection; So. Facade Unshaded False Color Luminance Image; values in candela/m
2

Chow | 57

4.1.7 SOUTH FAÇADE SHADED
The following images in Figure 4.6 are bracketed images taken at various exposure values. These images were taken after he
interior mechanized shades were lowered using an interior wall switch. Processed in Photoshop, the resulting .HDR File,
depicted in Figure 4.7, was converted to false color for analysis in hdrScope. It is important to note many features of the
interior are preserved due to the openness factor of the shade cloth. The differences between shade cloth photographs and
digitally rendered shade visualizations is discussed in Section 5.2.2.2.

Figure 4.7: Physical Data Collection; So. Facade Shaded HDR Composite Photograph

Figure 4.6: Physical Data Collection; South Facade Shaded, Bracketed Images taken with a Nikon D300
Chow | 58

The hdrScope software tool translated these images into false color luminance maps with labels describing peak luminance
data in the units candela/m
2
. This false color luminance map, depicts peak luminance along a scale determined by the
extreme values found in the image. A scale of 0–3000 candela/m
2
was specified for the map depicted in Figure 4.8. It is
important to note the peak luminance of the unshaded condition represents the maximum theoretical luminance value, as the
shaded condition effectively reduced luminance.

Figure 4.8: Physical Data Collection; So. Facade Shaded False Color Luminance Image, Values in candela/m
2

Chow | 59

4.2 PHASE 2: DIGITAL SIMULATION
4.2.1 ILLUMINANCE
Illuminance calculations simulated light levels at the interior glazing as well as the defined horizontal calculation planes. The
data outlined in Table 4.6, Table 4.7, and Table 4.8 were sourced utilizing Rhinoceros 5.0 and DIVA’s Radiance simulation
engine. The illuminance values recorded in the following tables are organized by horizontal calculation planes numbered 0 at
the ground plane, to 4 at the top of the façade construction. Calculation points evenly spaces along the window width are
noted by lowercase letters a through c, while uppercase letters A, B, and C describe points at the exterior façade face, 1 meter
and 2 meters from the façade, respectively. The metric used for the following measurements is the SI unit lux.
4.2.1.1 SOUTH FAÇADE UNGLAZED
Table 4.6: Data Processing; Illuminance Measurements South Facade Unglazed, Values expressed in lux
façade Illuminance
(lux)

At Int
Glazing
A B
(1m)
C
(2m)
Unglazed
3
a 12 10 5 1
b 10 8 4 1
c 9 14 2 1
2
a 16 18 4 1
b 14 19 4 1
c 18 22 5 2
1
a 26 24 9 3
b 27 24 7 2
c 29 24 6 5
(Ground) 0
a - 0 0 0
b - 0 0 1
c - 0 0 1

4.2.1.2 SOUTH FAÇADE UNSHADED
Table 4.7: Data Processing; Illuminance Measurements South Facade Unshaded
façade Illuminance
(lux)

At Int
Glazing
A B
(1m)
C
(2m)
Unshaded
3
a 15 8 2 0
b 18 7 3 0
Chow | 60

c 20 7 2 1
2
a 22 11 2 2
b 24 13 3 3
c 27 11 4 2
1
a 29 17 4 4
b 29 16 5 3
c 32 17 5 3
(Ground) 0
a - 0 0 3
b - 0 0 4
c - 0 0 3

4.2.1.3 SOUTH FAÇADE SHADED
Table 4.8: Data Processing; Illuminance Measurements South Facade Unshaded
façade Illuminance
(lux)

At Int
Glazing
A B
(1m)
C
(2m)
Shaded
3
a 12 0 0 0
b 14 0 0 0
c 12 0 0 0
2
a 21 1 0 0
b 24 1 0 0
c 20 1 0 0
1
a 33 1 0 0
b 28 1 0 0
c 23 1 0 0
(Ground) 0
a - 0 0 0
b - 0 0 0
c - 0 0 0

Chow | 61

4.2.2 LUMINANCE RENDERINGS
Luminance renderings were accomplished by running Rhinoceros and DIVA’s Radiance-based Visualization function. The
wxfalsecolor application utilizing evalglare resulted in luminance maps and the labeling of extreme luminance values in
candela/m
2
.
4.2.2.1 SOUTH FAÇADE UNSHADED
The Luminance rendering for the unshaded condition of the south façade was accomplished by exposing the glazing layer
and hiding the shade layer. Note the following Figure 4.9 label of extreme luminance values in candela/m
2
.

Figure 4.9: Digital Simulation; So. Facade Unshaded False Color Luminance Rendering,
Scale in candela/m
2

Chow | 62

4.2.3 SOUTH FAÇADE SHADED
The Luminance rendering for the shaded condition of the south façade was accomplished by exposing both the glazing layer
and the shade layer. Note the following Figure 4.10 labels of extreme luminance values in candela/m
2

Figure 4.10: Digital Simulation; So. Facade Shaded Luminance Rendering, Scale in
candela/m
2

Chow | 63

4.3 PHASE 3: REDUCTION POTENTIAL
4.3.1 ILLUMINANCE
Illuminance calculations simulated light levels at the interior glazing as well as the defined horizontal calculation planes for
all tested glazing scenarios using Rhinoceros 5.0 and DIVA’s Radiance simulation engine. The illuminance values recorded
in the following figures and tables are organized by horizontal calculation planes numbered 0 at the ground plane, to 4 at the
top of the façade construction. The tables plot the calculation planes on the Y axis, and illuminance values on the X axis.
Average plane illuminance values are connected by a thin line, with minimum and maximum values plotted adjacent to the
line. The tables list average, minimum, and maximum values for reference. Calculation points evenly spaced along the
window width are noted by lowercase letters a through c, while uppercase letters A, B, and C describe points at the exterior
façade face, 1 meter, and 2 meters from the façade, respectively. The metric used for the following measurements is the SI
unit lux.
4.3.1.1 UNGLAZED CONDITION
Table 4.9: Reduction Potential; Unglazed Illuminance Calculations
façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
Unglazed
Sky 4 - 0 0 0
Top Window 3 270 42 6.56 0
Mid Window 2 667 365 56.11 0
Lower Window 1 619 486 129.71 0
Ground 0 - 223 91.8 0

0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
U N G L A Z E D C O N D I T I O N
Chow | 64

4.3.1.2 SINGLE PANE CLEAR

Table 4.10: Reduction Potential; Single Pane Clear Illuminance Calculations
façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
Single Pane Clear
Sky 4 - 0 0 0
Top Window 3 257 35 5.99 0
Mid Window 2 668 307 48.87 0
Lower Window 1 619 399 110.75 0
Ground 0 - 181 76.64 0

76.64
110.75
48.87
5.99
0
0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
S I N G L E P A N E C L E A R
Chow | 65

4.3.1.3 DOUBLE PANE CLEAR

Table 4.11: Reduction Potential; Double Pane Clear Illuminance Calculations
façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
Double Pane
Clear
Sky 4 - 0 0 0
Top Window 3 254 31 5.36 0
Mid Window 2 360 274 43.68 0
Lower Window 1 680 353 98.79 0
Ground 0 - 161 68.14 0

68.14
98.79
43.68
5.36
0
0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
D O U B L E P A N E C L E A R
Chow | 66

4.3.1.4 DOUBLE PANE LOW E

Table 4.12: Reduction Potential; Double Pane Low E Illuminance Calculations
façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
Double Pane Low
E
Sky 4 - 0 0 0
Top Window 3 253 25 4.28 0
Mid Window 2 644 217 34.73 0
Lower Window 1 609 278 78.19 0
Ground 0 - 127 53.5 0

53.5
78.19
34.73
4.28
0
0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
D O U B L E P A N E L O W E
Chow | 67

4.3.1.5 TRIPLE PANE KRYPTON AIR GAP

Table 4.13: Reduction Potential; Triple Pane Krypton Gap Illuminance Calculations

façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
Triple Pane
Sky 4 - 0 0 0
Top Window 3 252 17 3 0
Mid Window 2 643 149 24.23 0
Lower Window 1 609 190 54.12 0
Ground 0 - 86 36.59 0
36.59
54.12
24.23
3
0
0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
T R I P L E P A N E
Chow | 68

4.3.1.6 SOLARGARD FILM

Table 4.14: Reduction Potential; SolarGard Film Illuminance Calculations
façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
SolarGard Film
Sky 4 - 0 0 0
Top Window 3 266 6 4.85 0
Mid Window 2 653 56 35.07 0
Lower Window 1 618 74 57.62 4
Ground 0 - 37 14.63 0

15.78
24.45
11.28
1.4
0
0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
S O L A R G A R D F I L M
Chow | 69

4.3.1.7 MECHOSHADE WITH 5% OPENNESS FACTOR

Table 4.15: Reduction Potential; Mechoshade Illuminance Calculations
façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
Mechoshade O.F. _5%
Sky 4 - 0 0 0
Top Window 3 286 5 0.63 0
Mid Window 2 681 24 3.18 0
Lower Window 1 632 25 6.49 1
Ground 0 - 10 3.83 0

4.58
7.7
3.82
0.77
0
0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
M E C H O S H A D E 5 % O P E N N E S S
Chow | 70

4.3.1.8 SOLARGARD FILM AND MECHOSHADE

Table 4.16: Reduction Potential; Shade and Film Illuminance Calculations
façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
Shade + Film
Sky 4 - 0 0 0
Top Window 3 286 1 0.13 0
Mid Window 2 680 3 0.64 0
Lower Window 1 633 4 1.36 1
Ground 0 - 2 0.78 0

0.78
1.36
0.64
0.13
0
0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
S O L A R G A R D + M E C H O S H A D E 5 % O P E N
Chow | 71

4.3.1.9 OPEN LOUVERED SHADE

Table 4.17: Reduction Potential; Louvered Open Shade Illuminance Calculations
façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
Louvered Shade IGU Open
Sky 4 - 0 0 0
Top Window 3 254 58 5.82 0
Mid Window 2 667 404 42.99 0
Lower Window 1 630 519 76.41 0
Ground 0 - 68 26.82 0

26.82
76.41
42.99
5.82
0
0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
L O U V E R E D S H A D E _ O P E N
Chow | 72

4.3.1.10 LOUVERED SHADE 45º

Table 4.18: Reduction Potential; Louvered 45 Deg. Shade Illuminance Calculations
façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
Louvered Shade IGU 45 Deg
Sky 4 - 0 0 0
Top Window 3 231 1 0.26 0
Mid Window 2 640 4 0.94 0
Lower Window 1 619 6 1.73 0
Ground 0 - 2 0.68 0

0.68
1.73
0.94
0.26
0
0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
L O U V E R E D S H A D E _ 4 5 °
Chow | 73

4.3.1.11 CLOSED LOUVERED SHADE 45º

Table 4.19: Reduction Potential; Louvered Closed Shade Illuminance Calculations
façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
Louvered Shade IGU Closed
Sky 4 - 0 0 0
Top Window 3 288 1 0.12 0
Mid Window 2 681 4 0.77 0
Lower Window 1 635 5 1.62 0
Ground 0 - 2 0.94 0

0.94
1.62
0.77
0.12
0
0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
L O U V E R E D S H A D E _ C L O S E D
Chow | 74

4.3.1.12 CELLULAR INSULATED GLAZING UNIT (IGU)

Table 4.20: Reduction Potential; Cellular IGU Illuminance Calculations
façade Illuminance
(lux)

Peak At Int
Glazing
Max Avg Min
Cellular IGU
Sky 4 - 0 0 0
Top Window 3 265 56 2.37 0
Mid Window 2 657 404 165.77 0
Lower Window 1 618 519 76.41 0
Ground 0 - 68 188.96 0

26.82
76.41
42.99
5.82
0
0
1
2
3
4
0 50 100 150 200 250 300 350 400 450 500
P L A N E
I L L U M I N A N C E ( L U X )
C E L L U L A R I . G . U .
Chow | 75

4.3.2 LUMINANCE
Both rendered visualizations and false color luminance maps were generated in DIVA, with maximum and minimum
luminance values labeled in candela/m
2
. The following figures present luminance maps alongside corresponding rendered
visualizations for comparison of luminance reduction and the estimated visual effect.
4.3.2.1 UNGLAZED
4.3.2.2 SINGLE PANE CLEAR GLASS
4.3.2.3 DOUBLE PANE CLEAR GLASS

Figure 4.11: Reduction Potential; Luminance Rendering of Unglazed Base Case, Scale in candela/m
2

Figure 4.12: Reduction Potential; Luminance Rendering of Single Pane Clear Glass, Scale in candela/m
2

Figure 4.13: Reduction Potential; Luminance Rendering, Double Pane Clear Glass, Scale in candela/m
2

Chow | 76

4.3.2.4 DOUBLE PANE LOW EMISSIVITY GLAZING
4.3.2.5 TRIPLE PANE KRYPTON GAP
4.3.2.6 SOLARGARD FILM

Figure 4.14: Reduction Potential; Luminance Rendering, Double Pane Low E, Scale in candela/m
2

Figure 4.15: Reduction Potential; Luminance Rendering, Triple Pane Glazing with Air Gap, Scale in candela/m
2

Figure 4.16: Reduction Potential; Luminance Rendering, Clear Glazing with Solar Control Film, Scale in candela/m
2

Chow | 77

4.3.2.7 MECHOSHADE 5% OPENNESS FACTOR
4.3.2.8 SOLARGARD FILM AND MECHOSHADE

Figure 4.17: Reduction Potential; Luminance Rendering, Clear Glazing with 5% Open Shade Cloth, Scale in candela/m
2

Figure 4.18: Reduction Potential; Luminance Rendering, Solar Control Film and Mechoshade Shade Cloth, Scale in candela/m
2

Chow | 78

4.3.2.9 LOUVERED BLIND

Figure 4.19: Reduction Potential; Luminance Rendering, Double Glazing 76mm Louvered Shade Open

Figure 4.20: Reduction Potential; Luminance Rendering, Double Glazing 76mm Louvered Shade 45° Orientation

Figure 4.21: Reduction Potential; Luminance Rendering, Double Glazing 76mm Louvered Shade Closed

Chow | 79

4.3.2.10 CELLULAR INSULATED GLAZING UNIT

Figure 4.22: Reduction Potential; Luminance Rendering, Double Glazing 12mm White Cellular Structure

Chow | 80

5 DISCUSSION
Chapter Five discusses the collection of real photometric data, the comparison of these results to simulated photometrics, and
the potential for various façade constructions to reduce interior light losses.
5.1 DISCUSISON OF DATA COLLECTION
5.1.1 CASE STUDY CONSIDERATIONS
As discussed in section 3.1, several important criteria were established for the selection of the case study building. The test
site accommodated many logistical concerns, including access to the building and its specifications. While this information
was easily negotiated with tenants, several design aspects of 800 Wilshire were not ideal for testing, and require discussion.
The open plan design did not represent a “typical” office layout for the commercial towers popularly found in Los Angeles’s
urban center. In fact, the recently renovated office spaces on the 16
th
floor adopted a contemporary open plan and exposed
ceiling which are starkly contrasted with conventional perimeter offices on floors below. The office/cubicle configuration
established in the 1960s was reinterpreted in the early 1990s to maximize floor area and inspire collaboration between
workers. Despite recent productivity studies which indicate increases in distractions among open office employees, the open
layout is undoubtedly popular. The International Facilities Management Association reports 70 percent of American workers
operate in open floor plan environments. Design firms continue to incorporate both open and private workstations in the
design of commercial interiors like those found in the Buro Happold office (Stuart 2016). The decision to test this space
considered the popular open and hybrid layouts as models for contemporary workspaces.
5.1.2 SURVEY
The results of the survey, as detailed in Section 4, were gathered during various visits to the Buro Happold Engineering
office. Often, visual inspection was sufficient to obtain these results. Measurements and paint color, for example, were easily
recorded on site. Some detailed information, however, was visually inscrutable. Window transmissivity values, for example,
could only be accurately obtained from manufacturer literature. These values were provided as part of industry standard
metrics describing the results of definitive laboratory testing. In similar cases, specification data was obtained through email
correspondence with Buro Happold engineers, material specifiers and shading manufacturers.
5.1.3 DATA COLLECTION MODULE
The use of the Data Collection Module preserved the fidelity of the light level measurements taken inside. As discussed in
Section 3.2.2, the dimensions of the module were influenced by information taken from survey measurements. The width of a
single exterior glazing unit constrained the span of the module to 1m. The distance from the façade to the 16
th
Floor terrace
guardrail limited the depth of the data collection module to 2m, allowing for three equidistant calculation planes. The overall
height of the testing module was limited by the length of pipe available for purchase at local home improvement stores, as
well as the interior limitations of the vehicle used for transport. Considering ease of transport and even the interior
dimensions of the office’s elevator cabs, it was decided that the longest pipe sections would be cut into segments and
assembled at the testing site. Though segmented into manageably-sized sections, the aggregation of pipes created an
organizational challenge. Therefore, a system was devised by which the module could be assembled efficiently without
risking the loss of parts. Pipes and fittings were pre-assembled and those supporting blackout fabric were threaded through
the sewn panels as rod pocket curtains. Adjacent pipe sections were gathered and rolled together in the blackout fabric. Each
bundle curtains and loose pipes could be assembled in minutes, allowing the remaining precious time devoted to testing.
After purchasing just the necessary sections of pipes, it was clear the material weight would become a significant constraint
for the size of the data collection module. For these experiments, the module was limited by the weight that could be
responsibly carried by one person. It must be noted that constraints for the module height concerning material size and weight
had a considerable effect on the results of the Photometric analysis. Limiting the height of the box decreased the total glazed
area available for testing inside the module. The result is the non-contributing region of the façade glazing unit, shaded in
blue in Figure 5.1. These considerations, intricately balanced with cost constraints, ensured the module could be erected
quickly and efficiently on testing evenings.

Chow | 81

5.1.4 PHOTOMETRIC ANALYSIS
The photometric analysis of 800 Wilshire was performed on two winter evenings: November 23rd, 2015, and December 11th,
2015. The opportunity to access the terrace after dark and within the normal operating hours of the Buro Happold office was
a significant advantage of performing the analysis during the winter season. Several additional
5.1.4.1 ILLUMINANCE
The photometric analyses performed at the 800 Wilshire South balcony façade revealed important phenomena pertaining to
façade illuminance mitigation. Light levels were measured at 48 individual collection points inside the Data Collection
module discussed in Section 3.2.2. The captured illuminance values are illustrated in Figure 5.2 and Figure 5.4. The applied
film was found to significantly reduce light leak, when comparing measured light levels at the glazing interior with those
measured at the glazing exterior. The observed reduction of illuminance during this phase suggested the effect of film-
mitigated light arriving incident a calculation planes 3 and 0 above and below the window area may not contribute
significantly to uplight and light trespass, respectively.
Figure 5.1: Diagram Showing the Data Collection Module at the Testing Façade,
Note the non-Contributing Portion of the Façade Unit Shaded in Blue
Chow | 82

Table 5.1: Physical Data Collection; Illuminance Measurements South Facade Unshaded
façade Illuminance
(lux)

At Int
Glazing
A B C
SolarGard Film
3
a 126 4 1 0
b 126 4 1 0
c 126 4 1 0
2
a 126 4 2 1
b 126 4 2 1
c 126 4 2 1
1
a 126 4 0 0
b 126 2 0 0
c 126 4 0 0
0
a - 0 0 0
b - 0 0 0
c - 0 0 0

5.1.4.2 LUMINANCE
Despite the expected inaccuracies inherent in the capture of physical light data, the cameral location was carefully measured
before photographs were taken. It is important to note the height of the camera was determined by the photographer’s eye
level, about 167.64 cm above the ground plane. Some negligible inconsistencies in the blackout fabric were visible in the
photographs due to displacement by wind conditions.
5.2 DISCUSSION OF DATA PROCESSING
5.2.1 DIGITAL MODELING CONSIDERATIONS
Because of the nature of lighting simulation, variables required assessment to produce results which were both accurate and
swift. The decision to model the case study conditions in a digital environment, rather than perform calculations for
luminance and illuminance enabled the simultaneous generation of photometric results and renderings. The 10
th
edition of the
Illuminating Engineering Society Lighting Handbook lists a number of considerations for this type of image-based analysis
(DiLaura and Illuminating Engineering Society of North America 2011). A similar framework was followed to promote
consistency with the accepted modeling methods of industry simulations. Because the modeling process involves the
Rhinoceros 5.0 Digital environment and the simulation engine DIVA, the listed considerations were handled by their relevant
hosts. The description of the testing environment, and simple definition of surface properties were handled in Rhinoceros,
while the definition of the light source properties, calculation of surface luminances and, transformation of surface
photometric or radiometric properties for display, and the image display were specified using the integrated DIVA user
interface.
5.2.1.1 VARIABLES IN RHINOCEROS
The description of the digital environment, was made simple by the results of the visual survey and measurements. The model
was generated using previous knowledge of modeling commands. To prepare a model suitable for simulation in DIVA, the
Figure 5.2: Location of Data Points within Data
Collection Module
Chow | 83

digital planes were modeled as simple, single surfaces with all directional vectors oriented to the space’s interior light sources.
The model surfaces were also divided into layers referencing their material properties so as to be easily specified using the
DIVA interface.
5.2.1.2 VARIABLES IN DIVA
Several important variables built into the DIVA plugin needed to be considered for the simulations. The following commands
found in the DIVA user interface describe the variables addressed for the simulations. The screen image in Figure 5.3Figure
5.3: Rhinoceros 5.0 DIVA Toolbar operates the following software options.

Figure 5.3: Rhinoceros 5.0 DIVA Toolbar
Location
DIVA’s simulation accuracy is dependent on accurate location data. Weather files in the conventional .epw file
format provide mean climate data sourced from weather stations which are oftentimes located far from testing sites.
Additionally, these files commonly consist of data from past decades which compound the potential for simulation
inaccuracy. To address these inaccuracies, a downtown Los Angeles weather data file was used. Recently generated
by Buro Happold Engineers for in-house energy metering and analysis, this building-specific weather data was
sourced from a station located conveniently on the roof of 800 Wilshire.
Nodes
The location of the testing nodes was based entirely on the points measured during physical testing. A summary of
the points can be seen in Figure 5.4: Horizontal Plane Calculation Nodes below. These nodes were oriented on an
imaginary horizontal plane, so as to capture the effect of the relevant vertical interior façade surfaces. Vertical
illuminance values, though useful when considering a façade plane as a light source, would have self-shaded the
photocell during physical testing.

Figure 5.4: Horizontal Plane Calculation Nodes
Chow | 84

Materials
The selection of material properties was imperative to simulating accurate results. Functions built into the DIVA
plugin allow for testing of specific finishes and light sources. The definitions of these materials were based on
manufacturer supported models and values when available.
Assign Materials
Layer-based material specification was accomplished using the DIVA Plugin. DIVA’s Thermal Materials,
used for simulating performance with respect to specific material properties were not used in the lighting
simulations. Instead, DIVA Daylighting Materials calculated relatively accurate results of surface
luminances with respect to pre-defined transmissivity and reflectance values.
Where materials in DIVA’s limited Radiance Materials Library did not meet the requirements of the
physical environment, custom materials were generated. This required analysis of the existing finishes and
simply-coded radiance materials added to the text-based library generated for each model by the plugin.
For use in DIVA, Radiance Glazing (GLASS) materials are dependent on transmissivity. Translucent
surface (TRANS) materials require values for transmitted specularity and roughness. Table 5.2 summarizes
the custom glazing materials generated for analysis. DIVA’s glass materials are governed by transmissivity,
which were calculated by manufacturer-supplied Transmittance values using a simple formula.

Table 5.2: Custom DIVA Materials
Data Processing
Name
Opennes
s Factor
(OF)
VLT
(Tv)
Trans
(tn)
Light Refl
Value(Rd)
Spec.
Reflect.
RGB Rough
Trans
Factor
(T trans)
Trans
Specularity
(Tspec)
BH_Mechoshade
(1300 series)
Int 0.05 0.11 - 0.8 0 0.70 0.01 0.12088 0.454545
BH_SolarGard
Film
Ext 0 0.23 0.25 0.34 0.34 - 0 0.40351 0
Int 0 0.23 0.25 0.17 0.17 - 0 0.575 0
Load IES Files
Concurrent with the development of this research, an update to the simulation plugin (DIVA 3.0) was
released. This release allowed for the importing of .ies files. This file extension refers to an ASCII
(delimited text) file containing photometric data for a particular luminaire. Since 1986, this has become the
most common file type in North America after its use was recommended by the IESNA standard LM-63-
83. This plugin upgrade simplified the process of uploading a manufacturer’s photometric file, potentially
increasing the accuracy of simulation results. Though cumbersome to replicate in the previous release of
DIVA, the definition of the light source properties allowed for the alteration of the fixture’s light loss
factor. A light loss factor is a multiplier commonly used in lighting design and engineering calculations to
simulate lumen depreciation. The selection of a light loss factor could not be specified when switching to
DIVA 3.0, which simulated the full power output of the specified fixture. The specified fixture, as noted in
the fixture schedule obtained by Buro Happold Lighting Engineers, was made available by the
manufacturer as an .ies file and incorporated into the digital model.
Dynamic Shading Controls
DIVA’s advanced shading control options provide static and variable shading options for simulating
interior shades without necessitating a geometry. The all-on/all-off Conceptual Dynamic Shading option
reflects all direct sunlight, allowing 25% of diffuse sunlight into a space. Detailed Dynamic Shading
options offer mechanical or newly-added electrochromic glazing controls with manual automated or
temperature-based operation. Thresholds for glare and light levels arriving upon a surface are important
parameters for these shading control options. Because the point-in-time experiments evaluated light levels
outside the range of practical thresholds these options were not utilized for testing.
Lighting Controls
Chow | 85

DIVA allows for the creation of simple and detailed lighting control groups. The control scenarios offered
include Manual On/Off Switching, Dimming and On/Off switching with occupancy sensors, and Dimming
with photocontrols. The lighting power and lighting setpoints are important factors for these simulated
controls. Because the point-in-time experiments evaluated lighting after dark hours, these control options
were not considered.
Metrics
DIVA’s Metrics Tab contains the main control options for various types of simulations. The various options for
Daylight Image, Daylight-Grid, and Thermal Single Zone Simulation controlled the overall environment parameters
and properties for image or data display(“Simulations in General” 2016).
Daylight Images
Found under the Daylight Image tab, the Visualization Simulation was the main operation used for
calculating luminance values and generating false color renderings. Important parameters for Image-based
Simulations remained consistent for all the images produced. The chosen Image Quality parameter was set
to High, which generated “reliable” results while hindering rendering speed. The selected Sky Condition
used a theoretical sunless option “Clear Sky Without Sun,” which disregards atmospheric light
contributions. The specified Date and Time recalled the date and time of physical data the collection:
December 11
th
at 22:00hrs in the designated mm-dd-hr format (12 04 22). The Camera Type was set to 180
degree Fisheye, which most closely replicates the images taken on site with a Nikon D300. The image size
was specified to 800x600 pixels, for rapid testing. These options also offered a false-color display option.
Daylight Grid-Based
The options found under the Daylight Grid-Based Simulation Categories contained the testing options
based on the creation of a node grid. Point-in-Time Illuminance was the main operation used for generating
illuminance values on a calculation plane specified by the physical data collection points. The Metric
selected was Illuminance, and the preferred unit selected was the SI unit Lux. Parameters indicating the sky
condition and Date and Time, were consistent with those selected for the Daylight Images Visualization
Simulations. Because the point-in-time experiments evaluated lighting after dark hours, the Daylight factor,
Climate-based, or Solar Radiation Simulations were not considered.
Thermal Single-Zone
DIVA’s powerful Thermal Analysis uses Energy plus to test specified daylighting and control strategies in
the modeled environment. In this simulation, Thermal materials are analyzed in conjunction with Occupant
Density, Equipment Power Density, Air Changes per Hour, efficiency factor for heating and cooling
systems, and heating and cooling set points.
5.2.2 DIGITAL MODEL VALIDATION
The results of digital simulation studies as reported in Chapter 4, require some discussion. This section discusses the
comparison of digital results to physical measurements taken during Phase 1 data collection, by which the digital modeling
process was validated as viable.

Chow | 86

5.2.2.1 SOUTH FAÇADE UNSHADED
Photometric analysis of the south façade resulted in illuminance values and luminance renderings and maps that were easily
compared to the HDR images taken during the data collection phase. Comparisons between images and renderings were
made between peak luminance values highlighted by hdrScope and Radiance software utilities, respectively. A comparison
between the photographed condition and the rendered visualization was made in Figure 5.5.
Illuminance values from Radiance-based simulations, illustrated in were recorded and compared with captured values from
physical testing. The values in the accompanying Table 5.3 represent the simulation accuracy as a percentage.

Figure 5.6: Diagrammatic Comparison of Physical (Left) and Digital Illuminance Measurements (Right)
Figure 5.5: South Façade Unshaded Condition HDR Compiled Image (Left) and Rendered (Right) Perspective Views from 1m
Chow | 87

Table 5.3: South Façade Unshaded Illuminance Measurements, Estimated Accuracy
façade Illuminance
(lux)

At Int Glazing A B C
Unshaded
3
a 11.90% 200.00% 200.00% -
b 14.29% 175.00% 300.00% -
c 15.87% 175.00% 200.00% -
2
a 17.46% 275.00% 100.00% 200.00%
b 19.05% 325.00% 150.00% 300.00%
c 21.43% 275.00% 200.00% 200.00%
1
a 23.02% 425.00% - -
b 23.02% 800.00% - -
c 25.40% 425.00% - -
(Ground) 0
a - - - -
b - - - -
c - - - -

Extensive variations between the measured and simulated illuminance values may have been the product of several modeling
considerations. Illuminance measurements in the digital office space were likely to exceed simulated measurements, as it was
modeled without the expected office equipment and clutter visible in the scene in Figure 5.5. Objects such as office chairs,
books, and computers, did not contribute to reductions in the model's simulated illuminances. Furthermore, the digital
luminaire model, discussed in Section 5.2.1.2, was expected to yield higher simulated light levels at full output, compared to
the installed luminaire which was subject to depreciation. To test this, furniture reflectance values were reduced to estimate
the effect of office clutter. The results summarized in Table 5.4 allow comparison between measured values on the left and
simulated values on the right. It is important to note the significant reductions and improved accuracy of the calculated
illuminance simulation results within 95% at the interior glazing plane.

Chow | 88

Table 5.4: South Façade Unshaded Illuminance Measurements (Left) Calculations (Right) With Reduced Furniture Reflectance Values,
Note Significant Reductions in Simulated Values (Right) for Furniture Light Reflectance Values
façade Illuminance
(lux)

façade Illuminance
(lux)

At Int
Glazing
A B C

At Int
Glazing
A B C
Unshaded Unshaded
3 3
a 126 4 1 0 a 120 4 1 0
b 126 4 1 0 b 120 3 1 0
c 126 4 1 0 c 120 3 0 0
2 2
a 126 4 2 1 a 120 5 2 2
b 126 4 2 1 b 120 5 3 3
c 126 4 2 1 c 120 5 4 2
1 1
a 126 4 0 0 a 120 7 1 0
b 126 2 0 0 b 120 4 1 0
c 126 4 0 0 c 120 4 1 0
(Ground) 0 (Ground) 0
a - 0 0 0 a - 0 0 0
b - 0 0 0 b - 0 0 0
c - 0 0 0 c - 0 0 0

Based on peak luminance values, the modeling process produced an accurate estimation of luminance values within 97.4%.
Combined in the following Figure 5.7 these values can be compared.

Figure 5.7: South Facade Unshaded Condition HDR False Color (Left) and Rendered (Right) Luminance Maps
Chow | 89

Though slight, the variation in the measured luminance may be due to several factors. The manufacturer supplied luminaire
model, discussed in 5.2.1.2, was expected to yield higher simulated light levels at full output, compared to the depreciated
installed luminaire. Variations in paint color and texture influenced peak luminance as well, considering both the digital and
real environment were lit indirectly with wall surface reflections. The high degree of accuracy confirmed the ability to
accurately model light levels in the Rhinoceros Digital environment and suggested luminance measurements would govern
the reduction potential of the façade systems.
5.2.2.2 SOUTH FAÇADE SHADED
Photometric analysis of the shaded south façade resulted in illuminance values and luminance renderings and maps that were
less easily compared to the HDR images taken during the Data Collection phase. Comparisons between images and
renderings were made between average luminance values highlighted by hdrScope and peak luminance values highlighted by
Radiance software utilities.
Illuminance values from Radiance-based simulations were recorded and compared with captured values from physical
testing. Combined in the following Table 5.5, these values can be compared.
Table 5.5: South Façade Shaded; Illuminance Measurements (Left) Calculations (Right)

Despite the high degree of accuracy achieved by simulating the unshaded condition, DIVA’s shaded façade simulation
resulted in highly disparate peak luminance results. Based on peak illuminance values, the modeling process produced an
accurate estimation of luminance values within 9.7%. The analysis of these results required an understanding of both the
physical shade cloth and the translucent material developed for this simulation as discussed in Section 5.2.1.2.
The analysis of the mechanized shade cloth, which transmitted interior night lighting through a 5% open basket weave,
resulted in high luminances through the portions of the cloth which were completely open as evident in both the luminance
renderings and the left luminance map in Figure 5.8. This was compared with the luminance map produced by DIVA’s
Radiance simulation on the right, which accounts for “openness” by evenly distributing light transmission across the modeled
surface.
façade Illuminance
(lux)

façade Illuminance
(lux)

At Int
Glazing
A B C

At Int
Glazing
A B C
Shaded Shaded

3 3

a 126 2 1 0 a 12 0 0 0

b 126 2 1 0 b 14 0 0 0

c 126 2 1 0 c 12 0 0 0

2 2

a 126 2 1 0 a 21 1 0 0

b 126 2 1 0 b 24 1 0 0

c 126 2 1 0 c 20 1 0 0

1 1

a 126 2 1 0 a 33 1 0 0

b 126 2 1 0 b 28 1 0 0

c 126 2 1 0 c 23 1 0 0

(Ground) 0 (Ground) 0

a - 0 0 0 a - 0 0 0

b - 0 0 0 b - 0 0 0

c - 0 0 0 c - 0 0 0

Chow | 90

The translucent material was the ideal Radiance definition for simulating the window as a source of diffused luminance
through a shading device with a defined openness factor, but was albeit incapable of rendering a completely accurate
visualization of the light loss phenomenon. With the understanding that the measured diffuse luminance level could never
exceed the simulated peak luminance of the digital model, the accuracy of the simulation process was determined by
comparing the average of the illuminated window region in the HDR image to the peak luminance highlighted by the map
(right) depicted in Figure 5.8.

Figure 5.8: South Facade Shaded Condition HDR False Color (Left) and Rendered (Right) Luminance Maps
Chow | 91

5.3 DISCUSSION OF REDUCTION POTENTIAL
The analysis of the reduction potential of façade shading systems was conducted using the modeling and simulation
processes validated in the Data Collection phase. The scene geometry was determined in Rhinoceros, and all lighting
considerations were defined in DIVA. The results of Reduction Testing, as reported in Section 4.3, were presented in terms
of the conducted photometric measurement for ease of comparison.
5.3.1 DIGITAL MODELING CONSIDERATIONS
5.3.1.1 VARIABLES IN RHINOCEROS
With the understanding that the case study building reflected a non-standard open office plan indirectly lit with an
unconventional lighting strategy, a second “shoebox model” was developed to test light level reductions. This model, illustrated
in Figure 5.9, incorporates conventional office spaces located on the lower levels of the same case study building. The square
boundary of the space was modeled spanning four façade glazed panels of similar dimensions to the case study building, for
an overall dimension of 3.7m. x 3.7m. The floor to ceiling height, in comparison to the 16th floor offices, was modeled lower
at 3.5m. As with the case study simulation model, the surface directional vectors were oriented to the interior light sources.
The model surfaces were organized by layer so as to be easily specified using DIVA’s material selection functions.

5.3.1.2 VARIABLES IN DIVA
Reduction testing was dependent on the accurate definition of digital shading devices for use in DIVA's Radiance
simulations. Diva4rhino.com and radiance-online.org were indispensable resources for understanding the work of industry
professionals and software experts simulating in Radiance.
Location
The same weather file required utilized during the data processing phase was also used during reduction testing. The
.epw file was generated by Buro Happold Engineers for in-house energy metering and analysis, and was sourced
from a weather station located conveniently on the roof of the 800 Wilshire building.
Nodes
The calculation planes utilized for the Reduction Testing phase were specified similarly to those modeled in the
Data Processing phase with few adjustments. An additional horizontal calculation plane was added at the upper
boundary of the tested façade geometry. This “sky” plane and the corresponding “ground” plane at the lower
boundary, were chosen to host calculation points which would determine direct light leak contributions to uplight
and light trespass, respectively. Between these planes, three calculation planes divided the façade glazing units
equally. These planes correspond to research referenced in Section 282.2 by Darula et al. in which the distinction
was made between the upper and lower halves of the window. As for the planes used in the Data Processing method,
the point grid distributed nodes evenly along each glazed façade unit width, but included a much denser net of points
in the array spanning 2m from the façade.

Figure 5.9: Digital Shoebox Model Reflecting Conventional Offices at 800 Wilshire
Chow | 92

Materials
The selection of material properties was imperative to simulating accurate reduction potential results. Table 5.6
summarizes the custom glazing materials generated for analysis. DIVA’s standard glass materials are included,
along with several other GLASS and TRANS materials developed for Reduction testing.
Table 5.6: Reduction Potential; Custom Radiance Materials for use in DIVA
Data
Processing

Name
Openness
Factor (OF)
VLT
(Tv)
Tran
(tn)
Light Ref
Value(R)
Spec.
Reflect.
RGB Rough
Trans
Factor
(T trans)
Trans
Specularity
(T spec)
BH_Mechosha
de (1300
series)
Int 0.05 0.11 - 0.8 0 0.7 0.01 0.1208 0.454545
BH_SolarGard
Film
Ext 0 0.23 0.25 0.34 0.34 - 0 0.4035 0
Int 0 0.23 0.25 0.17 0.17 - 0 0.575 0
Single Pane Clear - 0.88 0.96 0.2 0.2 0.96 - - -
Double Pane Clear - 0.80 0.87
Double Pane_Low E - 0.65 0.71
Triple Pane
-
Cellular Shade IGU 7mm Ø - 0.381 0.42
Louvered Shade IGU - 0.221 0.24
-
Custom Radiance Material Definitions
Though relatively simple to define manually, custom Radiance materials were “crosschecked” using
several free software tools developed by the Lawrence Berkeley National Laboratory. The use of
WINDOW 7.4 allowed the virtual construction of custom façade glazing systems via a friendly user
interface. The software analyzed daylight and heat transmission through complex shading systems well
outside the scope of night time analysis. The result was a Visible Light Transmittance factor that was
included in the glazing definition. A simple WINDOW construction illustrated in Figure 5.10, replicates a
cellular shade system using a manufacturer-supplied inner layer.

Chow | 93

OPTICS 6 also analyzed digital façade assemblies utilizing an extensive database of films and coatings,
resulting in highly complex optical features. OPTICS allowed custom façade constructions to be exported
as text-based Radiance material definitions, which could be utilized in various ways to simulate night light
loss through the façade.
IES Files
To adopt the more conventional lighting strategy used in 800 Wilshire’s lower level office spaces, a
commercial troffer was selected for use in reduction testing. A pair of Cooper Lighting’s 2’x4’2GCAML
Series fixtures was loaded into the Rhinoceros model for simulation.

Figure 5.10: WINDOW7.4 Interface; Panelite Clearshade Optical Data Results
Chow | 94

5.3.2 ILLUMINANCE
Reductions in illuminance were analyzed by comparing the average, maximum, and minimum lux values reported in Table
4.9. The values tested in the Phase 3 model were much larger than those achieved during Phase 2: Data Processing. This is
likely due to the proximity of the Phase 3 fixtures to the calculation grids and the cumulative effects of light losses through
the increased glazing area modeled for testing. The higher density node count allowed for clearer visualizations of light
propagation. DIVA’s simulation results include color coded cells at each node location which illustrate light intensity across
a range of illuminance values. The values are summarized in the following Figure 5.11.
Figure 5.11: Average Illuminance Values for Reduction Testing Conditions by Façade Treatment, Values in lux

91.8
129.71
56.11
6.56
0
0
1
2
3
4
0 20 40 60 80 100 120 140
P L A N E
I L L U M I N A N C E ( L U X )
A V E R A G E I L L U M I N A N C E B Y F A C A D E T R E A T M E N T
UNGLAZED CONDITION SINGLE PLANE CLEAR DOUBLE PANE CLEAR
TRIPLE PANE SOLAR GARD FILM MECHOSHADE
SHADE + FILM LOUVERED BLIND OPEN LOUVERED BLIND 45
LOUVERED BLIND CLOSED CELLULAR IGU
Chow | 95

5.3.2.1 ILLUMINANCE MAPS
Peak Illuminance values are colored in red for the following illuminance maps. Though purely graphic in nature, the maps
allow a viewer to understand patterns in the distribution of light loss outside the façade plane. For reference, the map on the
far left of Figure 5.12 illustrates the unglazed base case to which the other maps can be compared. The calculated maximum
and minimum illuminance results are also noted under each map Figure 5.12 combines the illuminance analysis of DIVA
standard glazing materials. These strategies reflect marginal reductions in light levels arriving at the horizontal calculation
planes. It is important to note the distribution of spill light exceeds the boundaries of the 2m calculation grid.

Figure 5.12: Reduction Potential Illuminance Maps for Unglazed Condition (Left) and various DIVA Glazing Materials
Chow | 96

For reference, the map on the far left of Figure 5.13 illustrates the unglazed condition to which the other maps can be
compared. The calculated maximum and minimum illuminance results are also noted under each map. Figure 5.13 combines
the illuminance analysis of custom Radiance GLASS and TRANS materials, generated during Phase 2: Data Processing.
These strategies reflect significant reductions in light levels arriving at the horizontal calculation planes. It is important to
note the distribution of spill light is visibly reduced for all three scenarios. The maps illustrated in Figure 5.13 illustrate the
reduction potential of the installed shade and film strategies which exceed other glazing materials in terms of illuminance. .

Figure 5.13: Reduction Potential Illuminance Maps for Unglazed Condition (Left), Solargard Film and Mechoshade Shade Cloth (Right)
Chow | 97

Figure 5.14 illustrates the illuminance reduction potential of an IGU with integrated louvered shades. The 76mm shade was
modeled and tested in Rhinoceros to test the effect of the geometry on light leak. The following figure shows the changes in
illuminance determined by different orientation of the louvers. It is important to note the minimal distribution of light at Plane
0 which describes direct contributions to light trespass. This is a direct response to the orientation of the louver geometry.

Figure 5.14: Reduction Potential Illuminance Maps for Unglazed Condition (Left), Louvered Blinds
Opened, Oriented at 45º, and Closed (Right)

Chow | 98

Figure 5.15 illustrates the illuminance reduction potential of an IGU with integrated cellular structure shades. For illuminance
testing, A custom material was generated using WINDOW 7.4 optics data. While the 7mm-diameter cells were modeled in
Rhinoceros and tested for illuminance, the differences in illuminance were minimal. The following figure shows the
illuminance results of the cellular shade with the custom glazing material. Is important to note the intense scattering of the
cellular material encountered at calculation nodes adjacent to the glazing plane and the characteristically shallow distribution
along the calculation grid. This is accounted for by refraction caused by the cell material.

Figure 5.15: Reduction Potential Illuminance Maps for Unglazed Condition
(Left), Insulated Glazing Unit with Cellular Structure (Right)
Chow | 99

5.3.3 LUMINANCE
DIVA simulation was used to test the potential for luminance reduction using façade shading materials. Measured using false
color maps rendered at 1 meter from the façade, the data show significant reductions in peak luminance. Figure 5.16 depicts
peak luminance by shading type as well as the corresponding reduction percentage.
Figure 5.16: Luminance Reduction at 1m from the Façade by Shading Type
5.3.4 RENDERINGS
The rendered results of DIVA’s Daylight Visualization simulations, which were used to generate luminance maps using the
wxfalsecolor utility illustrate a critical flaw in the simulation method which requires discussion. Much like the material
parameter assumptions discussed in Section 5.2.2.2, which prevented DIVA from accurately visualizing light loss through a
shade cloth, the definition of the cellular shade structure resulted in an unexpected visualization. Compared with the expected
visualization pictured in Figure 5.17, neither simulated image reflects a truly accurate image of the cellular shade structure.
This may suggest true luminance visualizations are unattainable using this method. To test the visualization of a cellular
structure, the specified cell diameter was modeled as a 3D geometry. The slow progress of the rendered visualization offered
a discouraging conclusion that DIVA’s simulation ability was not able to accurately visualize the complex shading at the time
of reduction testing.
2824.625
2506.755
2224.997
1808.87
1279.528
588.83
46.841
8.63
1143.639
165.911
37.241
250.402
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
90.00%
100.00% 0
250
500
750
1000
1250
1500
1750
2000
2250
2500
2750
Unglazed Single Pame
Clear
Double
Pame Clear
Double
Pame Low E
Triple Pane SolarGard
Film
Mechoshade
O.F. _5%
Shade +
Film
Louvered
Shade IGU
Open
Louvered
Shade IGU
45 Deg
Louvered
Shade IGU
Closed
Cellular IGU
PERCENT REDUCTION
PEAK LUMINANCE
cd/m
2
L U M I N A N C E R E D U C T I O N A T 1 M F R O M F A C A D E B Y S H A D I N G T Y P E
Chow | 100

Figure 5.17: Cellular Structure Visualization, Transmissivity Simulation (Left), Expected Visualization (Middle), Modeled Visualization
(Right); image credit: Panelite.us

Chow | 101

6 CONCLUSIONS
In Phase 1: Data Collection, the objective was to understand building sourced light loss at the commercial building scale with
quantitative measurements and a complete survey of applicable building features. To meet this goal, it was necessary to
identify an existing case study façade with the capability of being accessed at the window unit scale. An office tower at 800
Wilshire Blvd in Los Angeles was selected as a test building and analyzed using conventional measurement tools. With the
aid of a data collection module, a light meter was used to capture illuminance measurements. The same data collection
module was used with a digital camera and image editing software to capture luminance values from outside the building. To
understand the process of light loss simulations as a representation of the existing conditions, a digital process was developed
using accurate and accessible simulation tools. The Phase 2 Data Processing computational work was accomplished using
digital modeling and simulation software to test light loss in a digital environment which closely represented the case study
building at 800 Wilshire. Given the successful replication of light loss measurement values, the Phase 3: Reduction Testing
computational work investigated the simulation of light loss using conventional façade treatments. Reduction testing required
a second digital model, adapted from existing office spaces in the case study building, in which to test the effect of different
shading strategies on light leak. The estimated reduction potential of various typical commercial building façade
constructions were presented in Chapter 4.
6.1 CONCLUSION OF FINDINGS
The photometric analyses of 800 Wilshire, performed on November 23rd, and December 11th, 2015, suggest a high potential
for light pollution reduction with the use of various façade treatments. The results of data processing, within 97% accuracy
for illuminance measurements and within 80% of luminance measurements provided a strong foundation for the reduction
testing of simulated façade constructions.
6.1.1 WINDOW ILLUMINANCE
The photometric analyses of 800 Wilshire, performed on November 23rd, and December 11th, 2015, demonstrated a high
potential for light pollution reduction with the use of various façade treatments. The results of data processing, within 97%
suggested highly accurate simulations of reality were possible with the specified method and tools. Illuminance
measurements, which were initially thought to have a larger impact on the exterior environment, were substantially reduced
by various façade constructions. Reduction testing revealed that illuminance values were most strongly affected by film and
shading strategies, which effectively cut all light falling upon the horizontal ground plane. Negligible light levels were noted
2 meters from the building façade.
6.1.2 INTERIOR LUMINANCE
The primary concern of light exfiltration was substantiated by the luminance values that were measured across the façade
construction. Reduction testing in the digital environment revealed a Panelite IGU could significantly reduce luminance
values through the façade. Luminances were reduced by nearly 90% compared with an unglazed condition. This supports the
initial hypothesis that a cellular IGU could significantly reduce façade luminance.
6.1.3 DIRECT SKY GLOW
Geometry was not found to have an impact on light loss contributions to sky glow. Disregarding the contributions from light
reflecting off the ground plane (which are admittedly of great concern outside the digital environment), no façade shading
strategy had a measurable effect on uplight contributing to direct sky glow.
6.1.4 RESEARCH CONCLUSIONS
The findings which document the reduction of light levels due to façade treatments support the initial research hypothesis. It
was found that conventional façade shading strategies modeled in Rhinoceros 5.0 and analyzed in DIVA could have a
significant impact in reducing exterior light levels in the form of illuminance and luminance measurements. The reduction
potential of several façade shading systems may be used strategically by designers to effectively reduce urban sources of
illumination which contribute to light pollution. A stronger link to conventional light pollution metrics is needed to
definitively conclude reductions in building-sourced light leak may reduce light pollution as it is currently defined by light
pollution researchers.
In keeping with the scope of the performed experiments, measurements were taken in the architectural context, utilizing
metrics common to the building and engineering disciplines. The experiments were conducted at a manageable scale,
considering methods and source data consistent with industry standards. Utilizing accessible and affordable software and
Chow | 102

measurement tools, Illuminance and Luminance values were able to be accurately measured or simulated. Limited access to a
spectrometer as well as working knowledge of reliable software for reading spectral data from image files pushed spectral
analyses of light loss outside of the research scope. An integral part of the debates surrogating the dangers of light pollution,
analyses of spectral content is included as a relevant topic for future work. Ecological and medical consequences of light leak
were discussed, but no claims were made to suggest a façade’s ability to prevent disease caused by light pollution sources.
6.2 APPLICATION OF RESEARCH
6.2.1 CHANGING REGULATIONS
Since the development of state-wide green building standards in 2007, California legislation has enjoyed a connotation for
being exceptionally conscious of environmental degradation. Active support for environmentally conscious building practices
have made the state of California one of the greenest states in the country. As the state’s energy policy agency, the California
Energy Commission develops energy strategies that have various impacts on lighting. The adoption of California Green
Building Standards Code (CALGreen), and its lighting-specific Title 24 stipulations continue the momentum of higher
efficiency for interior and exterior illumination for the state.
Lighting requirements continue to evolve more stringent requirements. An increasing trend towards the use of lighting
control systems has a particular impact on environmental light propagation. While recent revisions of proposed language for
upcoming 2016 Title 24 lighting efficiency standards provide allowances for the compulsory installation of lighting controls
in existing nonresidential projects, the use of occupancy/vacancy sensors may effectively do its part to reduce after hours
light leak coming from commercial interiors.
This does not invalidate the importance of this thesis research. Rather, it makes a stronger case for the use of control
strategies in existing buildings which do not qualify code “triggers.” Regulations involving these thresholds for compulsory
adoption of controls has fluctuated in recent months during the development of code language. This research is also relevant
for buildings with continuous operations, or those sited near sensitive ecological or protected sites. The design of buildings in
brightly-lit city centers in or out-of-state may take advantage of the research findings. While light exfiltration constitutes an
environmental issue regardless of a building’s occupancy, a wide adoption of these measures may have a significant impact
on artificial sky brightness, glare, and light trespass. This research proposes an alternative to control-based light pollution
reduction and should be addressed and readdressed as urban Los Angeles becomes a living laboratory for light pollution
reduction.
6.2.2 INVESTED PROFESSIONALS
For lighting consultants and engineers, the nature of this research may seem straightforward. With access to the necessary
equipment and digital tools, the experiments described in Chapter 3 may be replicated with every new project analyses. The
findings discussed in Chapter 5 may provide interesting strategies for light pollution reduction at the plane of the façade. For
lighting designers looking to quickly estimate a potential reduction of light leak, a simple rule of thumb may help in the
discovery of possible shading solutions. Consequently, utilizing a rule of thumb may result in severe inaccuracies. As
discussed in Chapter 1, this research included a limited scope of lighting sources and strategies. Variations in light
distribution, material finishes, and interior geometry may impact the results presented in this research.
For clients, this work may be seen as cumbersome and expensive. Building owners and developers looking to achieve
voluntary light pollution requirements like LEED’s SSC8, may become discouraged with the amount of work necessary to
validate this single point. As a result, applicants may choose to look to other possible sources of LEED points. However, the
basis of this research may encourage the use of façade treatments to achieve reductions in light levels and meet similar
sustainable thresholds.
For invested professionals in other disciplines, this research may become an important resource for support. As discussed in
Chapter 2, the variety of metrics used by researchers to quantify light pollution is sometimes incompatible with those used
daily by lighting designers to validate light pollution reduction. Addressing ecological light pollution in terms understood by
building professionals may help increase opportunities for building-specific reduction strategies.
Students looking into this research may find a wealth of unexplored scenarios to be analyzed and tested. Innovations to
building materials and methods are being developed very quickly and have the potential to change the nature of this work.
Several of these unmapped options are listed below:

Chow | 103

Future Work
Students interested in bridging a gap between architectural performance analyses and light pollution metrics may
consider a variety of research trajectories.
Building Scale Analysis
Both new and existing buildings are at risk to contributing to sources of light pollution. The analyses of
varying building types and geometries may result in significant methods for limiting light loss within a set
of building typologies. The most important next step is to investigate building-wide contributions to light
pollution, and the expected contributions of adjacent building sites. Analyses can be accurately done using
the digital tools outlined in Chapter 3, and the conversion of these simple metrics to conventional light
pollution metrics developed by invested researchers.
Additional Façade Types
Despite the reliable results from Radiance-based DIVA simulations, several limitations discussed in
Section 5.3 revealed the inability for the software to accurately visualize light loss. Advancements to the
software will undoubtedly correct these restrictions. Analyses of various façade types incorporating
innovative cladding methods and materials may support new findings. Incorporating these variations with
façade geometry development and form finding may result in new “self shading” after dark techniques.
Façade Physical Testing
To achieve accurate results regarding the reduction potential of various façade systems, it is recommended
that interested students consider testing materials using accurate tools within a testing lab. The student may
be able to source façade materials, films, luminaires, cellular shades, and louvers with partnerships with
manufacturers. The development of a highly controlled environment would ensure the accuracy of these
measurements, and perhaps validate the results outlined in Chapter 4.
The Effect of Developing Lighting Technology
Rapid developments in lighting technology have resulted in new understanding of the nature of fixtures,
color temperatures, and color rendering. The effect of these and other factors may be explored by future
work.
Instantaneous Compliance Analysis
The generation of a scripting tool that calculates the probability of Title 24/LEED compliance, much like
the tool found in DIVA for daylighting analysis could change the way lighting consultants and building
owners approach compliance for voluntary Light Pollution restrictions. A simplified workflow can
incentivize the achievement of goals like LEED SSC8.
BIM Integration
Integrating the simulation process into a Building Information Modeling (BIM) workflow may be a helpful
next step for the development of this research. With all light fixtures and façade treatments modeled, a BIM
environment can effectively host light level measurements at a whole building scale. Information rich
families and manufacturer specific BIM light fixture models may be replaced easily, with simulations
generated automatically to test reductions in light levels. Incorporating a script for ease of iteration, would
allow trades in the AEC industry to make informed decisions affecting various design priorities. A tool
could be created to generate exfiltration profiles for building footprints, helping designers see a whole
building as a “luminaire” in 3 dimensions. When linking Building Information Models of neighboring
buildings and sites, designers could gain the ability to view and react to the cumulative effect of adjacent
light profiles.
6.2.3 WHOLE BUILDING ANALYSIS
The investigation at the façade unit scale is highly specific, particularly as a “Ground-Up” approach as discussed in Section
2. The value of this research comes as it expands outwards to include “Sky-Down” light pollution measurement methods.
Understanding commercial building-sourced light loss and the façade constructions which affect it may provide builders with
yet another design tool to produce iconic and responsible night skylines. This middle ground, achieved through whole
building analysis, would establish a building’s light loss footprint and highlight areas where light levels could be reduced. For
example, a multi-story mixed-use building could be subdivided by function, each with a different lighting profile according
to scheduling and specification. This analysis may be performed after each change in the façade or interior lighting strategies
incorporating all lights on or using anticipated scenarios. With anticipated light leak information, designers can selectively
clad the enclosure to maximize light loss reduction and minimize the cost of using catchall solutions. Keeping in mind access
to daylighting and the probability for glare, an informed designer can coordinate shading strategies responsibly. Increased
Chow | 104

coordination between trades is a trending business model in the AEC industry, and access to accurate information regarding
the benefits of shading strategies on night light leak may only increase sustainable day and night lighting practices.
6.2.4 VALUE OF THE (NIGHT) SKYLINE
It is important to note, that while eliminating light loss would solve many of the ecological, environmental, architectural, and
astronomical annoyances of building-sourced light pollution, the use of blackout shades would extinguish what contemporary
society has come to appreciate as a night-time wonder. Even when compared with a breathtaking view of the stars, an iconic
skyline can easily attract the attention of modern man’s phototropic vision. It has become an insuperable symbol of the
metropolitan lifestyle and a desirable view for those fortunate enough to live in the heart of a developed city. Therefore, it
must be concluded that an effective light leak strategy must consider views from the building interior, which must be
maintained to preserve the function of a glazed façade. This may be accomplished with films and shades which allow views
from the interior to an extent, as well as views to the interior as the dark night sky preserves contrast with even filtered or
shaded interior lighting.
It bears noting that shades and films may not constitute the definitive façade strategy for achieving maximum reductions in
light exfiltration while preserving views from the interior. Shades, while efficient light mitigators, effectively block views
from the interior. Significant reductions to luminance and illuminance displayed by the IGU systems suggest strong light
mitigation may be achieved while still allowing for views through portions of the façade. Continued engineering efforts may
result in a more efficient use of IGU materials which benefit daytime shading and night light exfiltration concerns. The
complex preservation of views may perhaps be the result of the search of a better shade structure.
6.3 CONCLUDING REMARKS
Interior night lighting, whether an indication of occupied or unoccupied spaces, creates iconic identities for commercial
buildings sharing an urban skyline. But this vista also conceals the ecological, cultural, and astronomical consequence of
insensitive lighting design. Various disciplines have identified the relationship of the built environment to light pollution,
offering motivations for reductions in light loss. Despite these motivations, cities have continued to brighten with damaging
effects. Designers require a toolset which is developing just as rapidly to combat these effects. The most logical tool is that
which receives initial consideration. Designers may see the building façade as a membrane which must be designed for
exfiltration of night lighting as much as infiltration of daylighting, with the possibility of shading strategies which are
sensitive to lighting conditions of the day and night. Accurate digital analysis of interior night light leak coupled with façade
information may be a useful tool for estimating necessary reductions. With research supporting the light leak reduction
potential of conventional façade treatments, designers can make influential decisions managing light for interior and exterior
environments, and ultimately dim the damaging effects of an inevitably bright future.

Chow | 105

BIBLIOGRAPHY
Aksamija, Ajla. 2013. Sustainable Facades: Design Methods for High-Performance Building Envelopes. 1. Aufl. Hoboken,
New Jersey: John Wiley & Sons, Inc.
Aubé, Martin, Johanne Roby, and Miroslav Kocifaj. 2013. “Evaluating Potential Spectral Impacts of Various Artificial Lights
on Melatonin Suppression, Photosynthesis, and Star Visibility.” PLoS One 8 (7).
Bellia, L., A. Pedace, and G. Barbato. 2014. “Indoor Artificial Lighting: Prediction of the Circadian Effects of Different
Spectral Power Distributions.” Lighting Research and Technology 46 (6): 650–60.
Boyce, Peter R. 2014. Human Factors in Lighting. Third edition. Boca Raton: CRC Press/Taylor and Francis.
https://libproxy.usc.edu/login?url=http://www.crcnetbase.com/isbn/9781439874899.
Cinzano, P, and F Falchi. 2012. “The Propagation of Light Pollution in the Atmosphere.” Monthly Notices of the Royal
Astronomical Society 427 (4): 3337–57.
Cinzano, P., and F. Falchi. 2014. “Quantifying Light Pollution.” Journal of Quantitative Spectroscopy and Radiative
Transfer, Light pollution: Theory, modeling, and measurements, 139 (May): 13–20.
Cinzano, P, F Falchi, and C.D Elvidge. 2001. “The First World Atlas of the Artificial Night Sky Brightness.” Monthly
Notices of the Royal Astronomical Society 328 (3): 689–707.
Cinzano, P., F. Falchi, C. D. Elvidge, and K. E. Baugh. 1999. “Mapping the Artificial Sky Brightness in Europe from DMSP
Satellite Measurements: The Situation of the Night Sky in Italy in the Last Quarter of Century,” May.
http://arxiv.org/abs/astro-ph/9905340.
Darula, Stansilav. 2013. “Windows in Buildings: Potential Source of Light Pollution.” In Proceedings of the International
Conference on Light Pollution Theory, Modelling and Measurements, 51–56. Smolenice, Slovakia.
http://gpi.savba.sk/GPIweb/ogg/ikohut/LPTMM/proceedings_LPTMM-2013.pdf.
DiLaura, David L, and Illuminating Engineering Society of North America. 2011. The Lighting Handbook: Reference and
Application. New York, NY: Illuminating Engineering Society of North America.
Garstang, R.H. 2004. “Mount Wilson Observatory: The Sad Story of Light Pollution.” Observatory 124 (1178): 14–21.
Ghosh, Tilottama, Sharolyn J Anderson, Christopher D Elvidge, and Paul C Sutton. 2013. “Using Nighttime Satellite
Imagery as a Proxy Measure of Human Well-Being.” Sustainability (Switzerland) 5 (12): 4988–5019.
Holzman, David C. 2010. “What’s in a Color?” Environmental Health Perspectives 118 (1): A22.
Illumination Engineering Society of North America. 2011. “Design Guide for Sustainable Lighting: An Introduction to the
Environmental Impacts of Lighting.” Design Guide IES DD-22-12. IESNA.
“International Dark-Skies Association.” 2015. International Dark-Sky Association. Accessed August 24.
http://darksky.org/our-work/.
Kahle, Logan Q., Maureen E. Flannery, and John P. Dumbacher. 2016. “Bird-Window Collisions at a West-Coast Urban
Park Museum: Analyses of Bird Biology and Window Attributes from Golden Gate Park, San Francisco:
e0144600.” PLoS One 11 (1).
Kránicz, Balázs, Zoltán Kolláth, and Viktória Gyutai. 2008. “Modeling Light Pollution.” PRZEGLĄD
ELEKTROTECHNICZNY 84 (8): 80–83.
Kuechly, Helga U, Christopher C.M Kyba, Thomas Ruhtz, Carsten Lindemann, Christian Wolter, Jürgen Fischer, and Franz
Hölker. 2012. “Aerial Survey and Spatial Analysis of Sources of Light Pollution in Berlin, Germany.” Remote
Sensing of Environment 126 (November): 39–50.
Kumaragurubaran, Viswanathan. 2012. “High Dynamic Range Image Processing Toolkit for Lighting Simulations and
Analysis.” ProQuest Dissertations Publishing. http://search.proquest.com/docview/1034738463.
“Lincoln Property Completes Renovation of 800 Wilshire.” 2015. L.A. Biz. November 10.
http://www.bizjournals.com/losangeles/news/2015/11/10/lincoln-property-completes-renovation-of-800.html.
Lockwood, G. W, R. D Floyd, and D. T Thompson. 1990. “Sky Glow and Outdoor Lighting Trends Since 1976 at the Lowell
Observatory.” Publications of the Astronomical Society of the Pacific 102 (650): 481–91.
Longcore, Travis, and Catherine Rich. 2004. “Ecological Light Pollution.” Frontiers in Ecology and the Environment 2 (4):
191–98.
Loss, Scott R., Tom Will, Sara S. Loss, and Peter P. Marra. 2014. “Bird–building Collisions in the United States: Estimates
of Annual Mortality and Species Vulnerability.” The Condor 116 (1): 8–23.
Luginbuhl, Christian B, Dan M Duriscoe, Chadwick W Moore, Angela Richman, G. Wesley Lockwood, and Donald R
Davis. 2009. “From the Ground Up II: Sky Glow and Near-Ground Artificial Light Propagation in Flagstaff,
Arizona.” Publications of the Astronomical Society of the Pacific 121 (876): 204–12.
Luginbuhl, Christian B, G. Wesley Lockwood, Donald R Davis, Kevin Pick, and Jennifer Selders. 2009. “From The Ground
Up I: Light Pollution Sources in Flagstaff, Arizona.” Publications of the Astronomical Society of the Pacific 121
(876): 185–203.
Machtans, Craig S., Christopher H. R. Wedeles, and Erin M. Bayne. 2013. “A First Estimate for Canada of the Number of
Birds Killed by Colliding with Building Windows.” Avian Conservation and Ecology 8 (2): 6.
Chow | 106

Memari, Ali M., ebrary, and Architectural Engineering Institute. Committee on Curtain Wall Systems. 2013. Curtain Wall
Systems: A Primer. Vol. no. 126. ASCE Manuals and Reports on Engineering Practice. Reston, Virginia: American
Society of Civil Engineers. https://libproxy.usc.edu/login?url=http://site.ebrary.com/lib/uscisd/Doc?id=10830450.
Mizon, Bob. 2012. Patrick Moore’s Practical Astronomy: Light Pollution : Responses and Remedies (2nd Edition). Springer
New York.
Oba, Naoko, Kohji Kawakami, Toshie Iwata, Takuji Uozumi, and Shunsuke Kohko. 2005. “Sky Glow Caused by the Spill
Light from Office Buildings.” Journal of Light & Visual Environment 29 (1): 19.
Panda, Satchidananda, and Frederick Marks. 2015. “Circadian Lighting for Health.” ASHRAE Journal 57 (11): 84–86.
Parkins, Kaitlyn L., Susan B. Elbin, and Elle Barnes. 2015. “Light, Glass, and Bird—Building Collisions in an Urban Park.”
Northeastern Naturalist 22 (1): 84–94.
Parks, Bob. 2014. “The IDA Smart Urban Lighting Initiative.” International Dark-Sky Association.
Rich, Catherine, Travis Longcore, and ebrary, Inc. 2006. Ecological Consequences of Artificial Night Lighting. Washington,
DC: Island Press. https://libproxy.usc.edu/login?url=http://site.ebrary.com/lib/uscisd/Doc?id=10161049.
Sanchez de Miguel, D. Alejandro. 2015. “Variacion espacial, temporal y espectral de la contaminacion lumınica y sus
fuentes: Metodologıa y resultados.” Madrid: Complutense University of Madrid.
http://www.researchgate.net/profile/Alejandro_Sanchez_de_Miguel/publication/280077947_Variacin_espacial_tem
poral_y_espectral_de_la_contaminacin_lumnica_y_sus_fuentes_Metodologa_y_resultados/links/55a6943b08aeb4e8
e6469ec8.pdf?inViewer=true&disableCoverPage=true&origin=publication_detail.
Schiler, Marc. 1992. Simplified Design of Building Lighting. Parker-Ambrose Series of Simplified Design Guides. New
York: Wiley.
“Simulations in General.” 2016. Accessed February 4. http://diva4rhino.com/user-guide/simulation-types/simulations-
general.
Soardo, P., P. Iacomussi, G. Rossi, and L. Fellin. 2008. “Compatibility of Road Lighting with Star Visibility.” Lighting
Research and Technology 40 (4): 307–22. doi:http://dx.doi.org.libproxy1.usc.edu/10.1177/1477153508092099.
Stuart, Hunter. 2016. “Companies Are Rethinking The Open Office, And It’s About Time.” The Huffington Post. Accessed
February 3. http://www.huffingtonpost.com/2015/02/12/open-offices-changing-to-include-private-
space_n_6669666.html.
Vincent, Roger. 2004. “Developer Buys Office Tower on Wilshire Blvd.” Los Angeles Times, March 3.
http://articles.latimes.com/2004/mar/03/business/fi-office3.
Weier, John. 2000. “NASA Earth Observatory :” Text.Article. October 23. http://earthobservatory.nasa.gov/Features/Lights/.
Wright, Maury. 2015. “IES Publishes TM-30 Defining New Color Metrics for Characterizing Lighting.” Ledsmagazine.com.
August 19. http://www.ledsmagazine.com/articles/2015/08/ies-publishes-tm-30-defining-new-color-metrics-for-
characterizing-lighting.html.
Zink, Robert M., and Joanna Eckles. 2010. “Twin Cities Bird-Building Collisions: A Status Update on ‘Project Birdsafe.’”
The Loon 82 (1): 34–37.

Abstract (if available)

Abstract In the discipline of building science, considerable effort is expended to evaluate the infiltration of light through a commercial building facade. The designer/engineer must mitigate daylight through glazing, and must combat issues of glare and internal heat gain. After dark, roadway and architectural lighting spill into the environment as urban centers become their own sources of illumination. The effect is increasing night sky brightness, glare, light trespass, and other light pollution phenomena at a global scale. With increasing attention to the effects of light pollution on the fields of astronomy, ecology, medicine, and design, Architects and Engineers may consider the alternative path of travel of light through a facade and imagine the degree of loss reduction due to infiltration mitigation strategies. To answer these and other questions, a methodology was developed to gather case study data from the 800 Wilshire Blvd. office tower in Downtown Los Angeles. To measure interior light loss through the building facade, luminance and illuminance were evaluated using an HDR-capable camera outfitted with a fisheye lens and an illuminance meter. These measurements were used to validate a digital model of the existing lighting conditions with which building-wide illuminance values and luminance renderings were generated. With the simulation method successfully validated, a digital “shoebox” model of a conventionally lit office space was created and an unglazed baseline condition generated. Various façade conditions were attributed to the model, including films, shades, an insulated glazing unit (IGU) with louvered blinds, and an integrated glazing unit with a cellular structure developed by Panelite. Simulation revealed reductions in illuminance by the cellular structure IGU, louvers oriented at 45°, and closed louvers, with significant reductions achieved with the use of shade cloth. Simulation renderings revealed luminance reduction of 60% by opened louvered blinds, 94% for louvered blinds oriented at 45°, and 99% for closed louvered blinds. Luminance reductions for the Panelite structure were estimated at 91% based on simulation results, though visualizations were inconclusive. Building managers and designers who contend with strict light pollution regulations may accept façade mitigation simulations as a source for support. By addressing light spill with compatible metrics and validation strategies, these professionals could make the case for built-in sensitivity to exfiltration. Understanding the role of the façade construction and the light pollution reduction potential of materials may increase these opportunities for building-specific light management strategies.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Daylighting in Riyadh and Los Angeles: comparison of cultural factors in potential market penetration

PDF

Multi-domain assessment of a kinetic facade: determining the control strategy of a kinetic façade using BIM based on energy performance, daylighting, and occupants’ preferences; Multi-domain asse...

PDF

Adaptive façade controls: A methodology based on occupant visual comfort preferences and cluster analysis

PDF

Integrating non-visual effects of lighting in the evaluation of electrical lighting designs for commercial offices

PDF

Evaluation and development of solar control screens: using daylight simulation to improve the performance of facade solar control screens

PDF

Facade retrofit: enhancing energy performance in existing buildings

PDF

Green facades: development of a taxonomy tool to assist design

PDF

Kinetic facades as environmental control systems: using kinetic facades to increase energy efficiency and building performance in office buildings

PDF

Human-building integration: Investigation of human eye pupil sizes as a measure of visual sensation in the workstation environment

PDF

Absolute glare and relative glare factors: predicting and quantifying levels of interior glare and exterior glare caused by sunlight and daylight

PDF

Energy use intensity estimation method based on building façade features by using regression models

PDF

Comparing visual comfort metrics for fourteen spaces using simulation-based luminance mapping

PDF

Skin fit and retrofit: challenging the sustainability of curtainwall practice in tall buildings

PDF

Double skin façade in hot arid climates: computer simulations to find optimized energy and thermal performance of double skin façades

PDF

Façade retrofit case study: the Edith Green-Wendell Wyatt Federal Building

PDF

Exterior glare simulation: understanding solar convergence from concave facades using heat maps

PDF

Performative shading design: parametric based measurement of shading system configuration effectiveness and trends

PDF

Natural ventilation in tall buildings: development of design guidelines based on climate and building height

PDF

Daylight prediction: an evaluation of daylighting simulation software for four cases

PDF

Comparative evaluation of lighting design software programs for daylighting in buildings

Asset Metadata

Creator Chow, Dennis Joseph (author)

Core Title Blind to light loss: evaluating light loss through commercial building facades as a contribution to urban light pollution

School School of Architecture

Degree Master of Building Science

Degree Program Building Science

Publication Date 08/02/2016

Defense Date 04/29/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag exfiltration,facade,glare,illuminance,light pollution,light trespass,lighting design,luminance,OAI-PMH Harvest,photometry,sky glow,window

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Noble, Douglas (committee chair), Longcore, Travis (committee member), Schiler, Marc (committee member)

Creator Email chowdj@usc.edu,djchow09@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-289931

Unique identifier UC11280608

Identifier etd-ChowDennis-4707.pdf (filename),usctheses-c40-289931 (legacy record id)

Legacy Identifier etd-ChowDennis-4707-0.pdf

Dmrecord 289931

Document Type Thesis

Format application/pdf (imt)

Rights Chow, Dennis Joseph

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA