GL-6

cover_7054_AASHTO 10/17/05 11:09 PM Page i
© 2005 by the American Association of State Highway and Transportation Officials.

All rights reserved. Duplication is a violation of applicable law.
American Association of State Highway and Transportation Officials
444 North Capitol Street, NW, Suite 249
Washington, DC 20001
P: 202-624-5800, F: 202-624-5806 fax
www.transportation.org

ISBN: 1-56051-325-X Publication Code: GL-6

AASHTO 2004–2005 Executive Committee
Voting Members
Officers:
President: Harold E. Linnenkohl, Georgia
Vice President: David Sprynczynatyk, North Dakota
Secretary-Treasurer: Larry M. King, Pennsylvania
Regional Representatives:
REGION I: Dan Tangherlini, District of Columbia, One-Year Term
Jim Capaldi, Rhode Island, Two-Year Term
REGION II: Harold Linnenkohl, Georgia, One-Year Term

Joe McInnes, Alabama, Two-Year Term
REGION III: Frank Busalacchi, Wisconsin, One-Year Term

Carol Malnau, Minnesota, Two-Year Term
REGION IV: David Sprynczynatyk, North Dakota, One-Year Term

Victor Mendez, Arizona, Two-Year Term
Nonvoting Members
Immediate Past President: Jack Lettiere, New Jersey
AASHTO Executive Director: John Horsley, Washington, DC
iii
Technical Committee on Highway Lighting
Chairman: Karl A. Burkett, P.E., Texas
Secretary: Balu Ananthanarayanan, P.E., Wisconsin
Carl Andersen, FHWA
Stanley C. Biddick, P.E., Alabama
Mike DeAngelo, New Jersey
Jon Dickinson, Nevada
Ernest Kim, P.E., Oregon
Susan Lodahl, P.E., Minnesota
Jeff Unick, P.E., Pennsylvania
iv
AASHTO Highway Subcommittee
on Traffic Engineering
Chair: Delbert McOmie Chair: Vacant

Wyoming
Vice Chair: Thomas Hicks Secretary: Regina McElroy
Maryland FHWA
Liaison: Ken Kobetsky
AASHTO
State Member
Alabama Timothy C. Taylor, P.E.
Alaska Kurtis J. Smith, P.E.
Arizona Mike Manthey, P.E.
Richard C. Moeur, P.E.
Arkansas Eric Phillips
Tony Sullivan
California Asif J. Haq
Karla Sutliff
Colorado Gabriela Vidal
Connecticut John F. Carey, P.E.
Delaware Donald D. Weber, P.E.
District of Columbia Douglas E. Noble, P.E.
Florida Lap Thong Hoang, P.E.
Georgia Keith Golden, P.E.
Hawaii Alvin Takeshita
Idaho Lance Johnson
Illinois Joseph S. Hill
Indiana Jim Poturalski
Carl T. Tuttle
Iowa Timothy D. Crouch
Kansas David A. Church, P.E.
Kenneth F. Hurst, P.E.
Kentucky Duane H. Thomas, P.E.
Louisiana Charles Adams, P.E., PTOE
Maine Bruce A. Ibarguen, III, P.E.
Maryland Thomas Hicks, P.E.
Massachusetts William Bent
Michigan Vacant
v
Minnesota Bernard J. Arseneau
Mississippi Wes Dean
Missouri Eileen Rackers
Montana Duane Williams, P.E.
Nebraska Randall D. Peters
Nevada Scott L. Thorson
New Hampshire William Lambert
New Jersey Douglas R. Bartlett
Timothy J. Szwedo
New Mexico Vacant
New York Bruce W. Smith, P.E.
North Carolina J. Kevin Lacy, P.E., CPM
North Dakota Al Covlin
Ohio Dave Holstein
Oklahoma Harold Smart
Oregon Edward L. Fischer, P.E., PTOE
Pennsylvania Glenn Rowe
Puerto Rico Samuel Forestier
Rhode Island Robert Rocchio, P.E.
South Carolina Richard B. Werts
South Dakota John Adler, P.E.
Tennessee Don L. Dahlinger
Gerald Gregory
Michael L. Tugwell
Texas Margaret (Meg) A. Moore
Utah Robert E. Hull
Vermont Vacant
Virginia Raymond J. Khoury, P.E.
Washington Theodore Trepanier, P.E.
West Virginia Barry Warhoftig, P.E.
Wisconsin David I. Vieth
Wyoming Michael N. Gostovich
Associate Member—Bridge, Port, and Toll

N.J. Turnpike Authority Robert F. Dale
Associate Member—Cities and Counties

City of Minneapolis Jon Wertjes
vi
Associate Member—Federal
USDA Forest Service John W. Bell
Associate Member—International
Manitoba Ben Rogers
Nova Scotia Ralph Hessian
Saskatchewan Sukhy Kent
vii
AASHTO Highway Subcommittee on Design
Chair: Allen D. Biehler Vice Chair: Robert L. Walters

Pennsylvania Arkansas
Secretary: Dwight A. Horne Liaison: Jim McDonnell
FHWA AASHTO
State Member
Alabama Don T. Arkle, P.E.

Steven E. Walker, P.E.
Alaska Gary Hogins, P.E.
Arizona Mary Viparina
Arkansas Phillip L. McConnell, P.E.
Charles D. Clements, P.E.
California Mark Leja, P.E.
Colorado Mithilesh “Mitch” Kumar
Gary W. Meacham
Connecticut Arthur W. Gruhn
Michael W. Lonergan
James H. Norman
Delaware Michael H. Simmons
Michael F. Balbierer
James M. Satterfield
District of Columbia Kathleen Penney
Zahra Dorriz
Allen Miller
Florida Brian A. Blanchard, P.E.
Robert Greer
Jim Mills, P.E.
Georgia James “Ben” Buchan, P.E.
Babs Abubakari, P.E.
Brent Story, P.E.
Hawaii Gary C.P. Choy
Julius Fronda
Idaho Loren D. Thomas
Steven C. Hutchinson
Illinois Michael Hine
Indiana Gary Mroczka, P.E.
Jeff Clanton, P.E.
Iowa Michael J. Kennerly
David L. Little
Deanna Maifield
viii
Kansas James O. Brewer, P.E.
Richard G. Adams, P.E.
LaMonte C. Armstrong, P.E.
Kentucky Kenneth Sperry, P.E.
David Jones
Louisiana N. Kent Israel
Nicholas Kalivoda, III
Lloyd E. Porta, Jr.
Maine Jerome A. Casey, P.E.
Maryland Kirk G. McClelland
Robert D. Douglass
Massachusetts John Blundo, P.E.
Stanley Wood, Jr.
Michigan Mark A. Van Port Fleet, P.E.
Minnesota Mukhtar Thakur, P.E.
Mississippi David Foster
John B. Pickering, P.E.
C. Keith Purvis, P.E.
Missouri David B. Nichols
Montana Paul R. Ferry
Lesly Tribelhorn
Nebraska Don Turek
Dawn Allyn
James J. Knott
Nevada Frank Csiga, Jr., P.E.
Ruedy Edgington
Rand Pollard, P.E.
Paul K. Sinnott, P.E.
New Hampshire Craig A. Green
New Jersey Kiran B. Patel
Brian Strizki
New Mexico Roy Maestas, P.E.
Dennis Peralta, P.E.
Max E. Valerio, P.E.
New York Philip J. Clark, P.E.
Daniel D’Angelo, P.E.
Richard W. Lee, P.E.
North Carolina Jay A. Bennett, P.E.
Deborah M. Barbour
Art McMillan
North Dakota Mark Gaydos
Ohio Cash Misel
Dirk Gross
Timothy McDonald
ix
Oklahoma Christine M. Senkowski, P.E.
Bruce E. Taylor
Oregon Thomas Lauer
Pennsylvania Dean A. Schreiber, P.E.
Brian D. Hare, P.E.
Puerto Rico Ariel Pérez
José E. Santana-Pimentel
Rhode Island J. Michael Bennett, P.E.
South Carolina Robert I. Pratt, P.E.
Rocque L. Kneece, P.E.
John V. Walsh, P.E.
South Dakota Joel Gengler
Joe J. Feller
Tennessee Jeff C. Jones
Mark Holloran
Texas Mark A. Marek
Utah Stan Burns, P.E.
Vermont Kevin Marshia, P.E.
Virginia Mohammad Mirshahi, P.E.
Barton A. Thrasher, P.E.
Washington Harold Peterfeso, P.E.
Richard Albin, P.E.
Ken L. Smith, P.E., CVS
West Virginia Jason C. Foster
Randolph T. Epperly, Jr.
Wisconsin Beth Cannestra
Wyoming Paul P. Bercich
Tony Laird
U.S. DOT Member

FAA Rick Marinelli, P.E.
Associate Member—Bridge, Port and Toll

N.J. Turnpike Authority J. Lawrence Williams
Penn. Turnpike Barry L. Troup, P.E.
Commission
Associate Member—Bridge, Port and Toll

Port Authority of NY and NJ Scott D. Murrell
Associate Member—Cities and Counties

City of Minneapolis Fred Abadi
x
Associate Member—Federal
USDA Forest Service Ellen G. LaFayette
Associate Member—International
Alberta Allan Kwan
British Columbia Richard Voyer
Ontario Joe Bucik
Ontario Joe Bucik
Saskatchewan Sukhy Kent
xi
Table of Contents
Chapter 1: Introduction ................................................................................................................1

Overview ......................................................................................................................1
Need for Engineering Expertise ...................................................................................1
Necessary Expertise .....................................................................................................1
Recommended Design Methods...................................................................................2
Modern Controls and Master Lighting Plans ...............................................................2
Chapter 2: Master Lighting Plans................................................................................................3

2.1 Overview ...........................................................................................................3
Introduction .......................................................................................................3
Definition...........................................................................................................3
Benefits..............................................................................................................3
2.2 Plan Development .............................................................................................3
Introduction .......................................................................................................3
Participants ........................................................................................................4
Groups Having Concerns About Lighting.........................................................4
2.3 Goals of a Master Lighting Plan........................................................................5
Introduction .......................................................................................................5
Improved Safety ................................................................................................5
Environmentally Judicious Use of Resources ...................................................5
Energy Use ........................................................................................................6
Tourists, Businesses, and Nighttime Activities.................................................6
Planned Maintenance ........................................................................................6
2.4 Lighting Curfews...............................................................................................6
Introduction .......................................................................................................6
Reasons for Curfews .........................................................................................7
Considerations Before Implementation.............................................................7
2.5 Electrical and Lighting Management Systems..................................................8
Introduction .......................................................................................................8
Cost....................................................................................................................8
Benefits..............................................................................................................9
Benefits of ELMS Standardization..................................................................10
2.6 Conducting Studies..........................................................................................10
Introduction .....................................................................................................10
Electrical System.............................................................................................11
Benefits and Effects of Curfews and Dimming...............................................11
Traffic Studies .................................................................................................11
Community Goals............................................................................................11
Budget .............................................................................................................11
Chapter 3: Techniques of Lighting Design................................................................................13

3.1 Overview .........................................................................................................13
Introduction .....................................................................................................13
Level and Uniformity of Illuminance and Luminance....................................13
Illuminance Considerations.............................................................................14
xiii
Roadway Lighting Design Guide
Luminance Considerations ..............................................................................14

Visibility-Based Design Methods....................................................................16
3.2 Warranting Conditions.....................................................................................17
General.............................................................................................................17
Benefits of Lighting.........................................................................................17
Rural Interchanges ...........................................................................................17
Warranting Conditions.....................................................................................17
Continuous Freeway Lighting .........................................................................18
Partial Interchange Lighting ............................................................................19
Special Considerations.....................................................................................20
Bridges.............................................................................................................20
3.3 Design Values for Freeways............................................................................20
General.............................................................................................................20
Design Considerations .....................................................................................20
Lighting Levels on Crossroads ........................................................................20
Partial Interchange Lighting ............................................................................21
Adaptation (Transition) Lighting.....................................................................22
Bridges and Overpasses...................................................................................22
Other Considerations .......................................................................................22
Area Classifications .........................................................................................22
3.4 Streets and Highways Other Than Freeways
(Including Walkways and Bicycle Ways) .......................................................23
General.............................................................................................................23
Warranting Conditions.....................................................................................23
Lighting Design Levels....................................................................................23
Other Considerations .......................................................................................23
3.5 Pole Placement Guidelines ..............................................................................26
Introduction......................................................................................................26
Obstruction of View ........................................................................................26
Height Restrictions ..........................................................................................26
Medians............................................................................................................26
Gore Areas .......................................................................................................26
Maintenance Considerations............................................................................26
Adjacent to Deflecting Barriers .......................................................................27
General Safety Concerns .................................................................................27
Design Options ................................................................................................27
Option 1: Remove the Obstacle.................................................................27
Option 2: Redesign the Obstacle ...............................................................27
Option 3: Relocate the Obstacle ................................................................28
Option 4: Use Appropriate Breakaway Device .........................................28
Option 5: Shield the Obstacle ....................................................................29
Option 6: Delineate the Obstacle...............................................................29
3.6 References........................................................................................................29
Chapter 4: Techniques of Lighting Design................................................................................31
Chapter 5: Tunnels and Underpasses ........................................................................................33

5.1 Overview .........................................................................................................33
xiv
5.2 Underpasses.....................................................................................................33
General Scope and Guide Application ............................................................33
Warrants for Nighttime Underpass Lighting...................................................33
Design Values for Underpass Lighting ...........................................................34
Selection and Placement of Underpass Luminaires ........................................34
5.3 Vehicular Tunnels ...........................................................................................34
General Scope and Guide Application ............................................................34
Short Tunnels ..................................................................................................35
Long Tunnels...................................................................................................35
Warrants for Tunnel Lighting..........................................................................35
Visibility Optimization of the Tunnel and Approach Features .......................36
5.4 Daytime Lighting of Tunnel Interiors .............................................................37
Short Tunnels—Silhouette Visibility ..............................................................37
Entrance Portal Lighting .................................................................................37
Lighting Beyond the Entrance Zone ...............................................................37
Nighttime Tunnel Lighting..............................................................................38
Selection and Placement of Tunnel Luminaires..............................................38
Tunnel Lighting Control Systems ...................................................................38
Maintenance Factor for Tunnel Lighting Design ............................................38
Chapter 6: Work Zone Lighting and Temporary Roadway Lighting....................................39

Introduction ................................................................................................................39
Cost.............................................................................................................................39
Types ..........................................................................................................................39
Design Values ............................................................................................................39
Safety..........................................................................................................................39
Work Area Lighting ...................................................................................................40
Chapter 7: Roundabouts .............................................................................................................41

Introduction ................................................................................................................41
Warrant Conditions.....................................................................................................41
Recommended Lighting Level ...................................................................................41
Chapter 8: Electrical Systems Requirements............................................................................43

Disconnects.................................................................................................................43
Guidelines Available ..................................................................................................43
Grounding...................................................................................................................43
Voltage Drop ..............................................................................................................43
Chapter 9: Safety Rest Areas......................................................................................................45

Overview .........................................................................................................45
Warrant............................................................................................................45
Design Values..................................................................................................45
Entrance and Exit ............................................................................................45
Interior Roadways ...........................................................................................46
Parking Areas ..................................................................................................47
Activity Areas..................................................................................................47
Main Lanes ......................................................................................................47
xv
Chapter 10: Roadway Sign Lighting..........................................................................................49

10.1 Overview .....................................................................................................49
Introduction .................................................................................................49
Key Elements of Roadway Sign Lighting...................................................49
10.2 Sign Lighting Warrants ...............................................................................50
10.3 Illuminated Sign Types ...............................................................................51
Ambient Luminance Classifications ...........................................................51
10.4 Sign Lighting Recommendations ................................................................52
Lighting Uniformity ....................................................................................53
Sign Color Standards...................................................................................53
Placement of Lighting Units........................................................................53
10.5 References ...................................................................................................54
Chapter 11: Maintenance Considerations in Roadway Lighting Design ...............................55

General......................................................................................................................55
Maintenance Factors .................................................................................................55
Luminaire Dirt Depreciation (LDD).........................................................................55
Lamp Lumen Depreciation (LLD)............................................................................55
Equipment Factor (EF) .............................................................................................56
Support Structure Maintenance ................................................................................56
Electrical Distribution and Control System Maintenance ........................................56
External Factors ........................................................................................................57
Chapter 12: Sky Glow and Light Trespass ...............................................................................59

12.1 Overview .....................................................................................................59
Introduction .................................................................................................59
Types of Objectionable Roadway Lighting.................................................59
12.2 Mitigating Sky Glow and Light Trespass....................................................60
Glossary ...........................................................................................................................................61
xvi
SCOH Ballot Draft 1–05
xvii
Chapter 1
Introduction
Overview
This guide replaces the 1984 publication entitled, An Informational Guide for Roadway Lighting. It
has been revised and brought up to date to reflect current practices in roadway lighting. The guide pro-
vides a general overview of lighting systems from the point of view of the transportation departments and
recommends minimum levels of quality. The guide incorporates the illuminance and luminance design
methods, but does not include the small target visibility (STV) method.
Two appendices to this document are located online for reference purposes at http://downloads.trans-
portation.org/lighting.pdf. Appendix A is entitled Literature Review and References, and contains infor-
mation on research studies related to accidents and highway lighting, driver parameters, pavement
parameters, and a summary of further references. Appendix B contains an overview of Lighting Basics,
including background information on the various issues related to effective highway lighting.
Need for Engineering Expertise
Most states require that final design documents be signed and sealed by a registered professional engi-
neer. The registrant is normally required to only sign and seal documents that the registrant prepared, or
documents for which the registrant was responsible for the direction and control of the work. Documents
include specifications, reports, drawings, plans, design information, and calculations. Lighting designs,
as described in this guide, meet the criteria for requirement of an engineering seal. This is necessary
because the public interest is at stake. In addition, the designs integrated with other aspects of the trans-
portation facility that have engineering features. These other aspects include support structures, break-
away devices, pavement characteristics, electrical characteristics, traffic engineering features, traffic
management features, and the relative priority of lighting with respect to other safety features on the facil-
ity. Proper engineering judgment must be used in the integration of the various aspects of the facility, as
well as in the consideration of maintenance and life-cycle costs.
Many of the current problems in outdoor lighting are due to poor lighting designs. These problems
include less than optimal benefits from the lighting systems, excessive glare and visual clutter at night,
the use of improper deprecation factors and photometric distributions, excessive sky glow and light tres-
pass, and excessive maintenance and energy costs. The primary cause of poor lighting designs is a lack of
proper lighting and electrical education.
Necessary Expertise
Departments of transportation (DOTs) that hire consultant engineers are paying for engineering exper-
tise. This expertise is available in the area of roadway lighting and associated electrical systems. The
expertise required for DOT lighting designs includes:
• lamp types and characteristics, including depreciation factors
• ballast types and characteristics
• fixture mechanical characteristics
1
• lens types
• photometric performance of luminaires and factors impacting such performance
• fixture mounting types
• pole mechanical and electrical characteristics
• breakaway device options and when appropriate to use
• clear zone criteria
• pole types, mounting options, and loading considerations
• foundation and support details
• pavement reflection factors
• mounting height and spacing options
• light trespass and sky glow issues including laws and ordinances
• lighting quality requirements, such as illuminance, luminance, veiling luminance, and visibility
• maintenance considerations for individual components and the lighting system as a whole
• energy and life-cycle costs
• coordination with master lighting plans.
Consultants accepting payment for lighting system design should perform that work in an engineering
manner. This includes exercising engineering judgement when balancing all of the above characteristics.
Training courses are available for engineers wishing to learn how to design lighting systems.
Recommended Design Methods
This guide recommends only the luminance and illuminance design methods but recognizes efforts to
turn lighting designs toward visibility-based design methods. A discussion of visibility-based methods is
included. The fact that this guide does not recommend the small target visibility (STV) design method
should not be interpreted as non-support for visibility based methods and associated research. The STV
method is not recommended as standard practice because the increased benefits of using this design
method have not been adequately demonstrated and confidence in the ability to achieve design values and
better visibility is low. Accurate visibility calculations require the input of all light sources, including off-
roadway sources and headlamps from multiple vehicles. Such data are not easily obtained on roadways
under even moderate traffic conditions. The fact that lighting is installed on roadways based on, among
other things, high traffic volumes lends further uncertainty to the value of visibility calculations. STV or
other visibility criteria may be applied as an additional quality criteria in order to gain experience with
those methods and to determine the value of such criteria.
Modern Controls and Master Lighting Plans
This guide also addresses modern control methods for lighting systems and encourages the use of a
master lighting plan incorporated into the traffic management centers, emergency management centers,
or other central locations. Better use of resources can be achieved through innovative strategies such as
lighting curfews, special event and weather situation control, and maintenance automation. Curfews can
create energy savings. Energy cost savings on one particular lighting system may allow additional light-
ing systems to be installed and operated in peak traffic times, thereby improving safety of the entire area
under the jurisdiction of the master lighting plan.
Maintenance inventory and management can be improved by remotely monitoring percentage of lights
burning, diagnosing problems prior to sending out maintenance crews, tracking life-cycle costs, and mon-
itoring the performance level of maintenance contractors.
2
Chapter 2
Master Lighting Plans
2.1 OVERVIEW
Introduction
The commitment to lighting roadway facilities is a large responsibility. A master lighting plan can help
in the fulfillment of this responsibility. This chapter provides general guidelines for implementing a mas-
ter lighting plan when local authorities determine that such a plan is desirable for their community.
Definition
A master lighting plan is a formal arrangement between local governments and other entities within a
regional area to coordinate and standardize the design, operation, and maintenance of public lighting.
Master lighting plans can include lighting curfews and sophisticated monitoring systems (described in
Sections 2.4 and 2.5, respectively).
Benefits
The basic benefits of lighting include safety, beautification, and security for people and property.
Additional benefits derived from a master lighting plan include:
• improved safety through the maximizing of resources
• a consistent image, reflecting the local culture and tastes
• nighttime linking of various sections of the city
• systems that better identify the nature of the site (residential versus “restaurant row,” for example)
• better management of energy use
• tighter control of sky glow and light trespass
• aid in implementing lighting curfews
• increased public security (other concerns may warrant immediate turning on or off)
• coordinated maintenance
• easier coordination of maintenance specifications, such as poles, breakaway devices, and luminaires.
2.2 PLAN DEVELOPMENT
Introduction
The master lighting plan development process can proceed as shown in the following table.
3
TABLE 2-1. Master Lighting Plan Development Process
Step/Action More Information
1. Coordinate with other participants to set goals. See following heading for information on participants.
See Section 2.3 for information on goals.
2. Consult with and consider the concerns of See “Groups Having Concerns About Lighting”
various groups having a stake in public later in this section.
lighting.
3. Conduct a study to justify and determine the See Section 2.3 for details on lighting curfews.
feasibility of the planned strategies. See Section 2.4 for details on electrical and lighting
management systems.
See Section 2.5 for details on conducting studies.
Participants
The master lighting plan approach is a regional concept that could involve the following entities:
• the state DOT
• city
• county
• police
• traffic management centers, emergency management centers, or other central locations
• fire, EMS, and other emergency agencies
• tunnel operators
• chambers of commerce (event coordination)
• parks (bike and pedestrian trails, fireworks, sports fields, etc.)
• the Federal Aviation Administration (near airports)
• the Coast Guard (near waterways)
Coordination Needed. Officials of these organizations can coordinate all publicly owned lighting sys-
tems by setting joint goals and laying out plans to achieve those goals. A master lighting plan should be
seen as a component of city management.
Groups Having Concerns About Lighting
In addition to the participating entities, other groups may have concerns related to public lighting, and
their concerns should be considered. These groups may include:
• citizens and property owners
• retailers and other business
• tourists
• visitors
• parks and sports facilities (including spectator sports and participant sports such as driving ranges
and softball)
4
2.3 GOALS OF A MASTER LIGHTING PLAN
Introduction
A master lighting plan should have five major goals:

• improved safety
• environmentally judicious use of resources
• judicious energy use
• attracting tourists, businesses, and nighttime activities
• planned maintenance.
Discussions of each of these goals follow under separate headings.
Improved Safety
Improved safety is the primary goal of public lighting. Public lighting affects motorists, cyclists, and
pedestrians. Public lighting includes roadways, sidewalks, and signs. A master lighting plan can help
maximize available resources and increasing management coordination. Maximizing resources means
lowering the operating costs of existing and new lighting systems, which may allow the installation and
operation of more lighting systems that operate only when needed.
Environmentally Judicious Use of Resources
“Environmentally judicious use of resources” refers to the consideration of safety, security, aesthetic
and architectural compatibility, creature comfort, sky glow, light trespass, glare, and visual clutter.
Consideration should be given to the desires of the facility owner, city officials, and citizens as
expressed in public hearings, local laws and ordinances, and other sources.
All outdoor lighting creates some degree of sky glow and light trespass. Modern controls can imple-
ment switching or dimming curfews that better address the needs and desires of the community (see
Section 2.4 for more information). In short, all community and business desires should be respected, and
competing concerns should be balanced as much as possible.
A wide variety of alternatives are available for outdoor lighting. Various environmental factors should
be subjectively weighted for different situations. Some factors to consider include:
• purpose of the lighting
• source type
• degree of light cutoff
• light level
• mounting height
• ambient light levels (other lighting in the area)
• historical considerations
• area land use
• roadway classification
• pedestrian and cyclist use
• implementation of lighting curfews (see Section 2.4)
• effect on wildlife (some studies show that nighttime lighting may affect wildlife)
A review of the following facility types should demonstrate an intuitive weighting:
• urban freeways
• rural freeways
• central business district (CBD)
5
• arterial roadways
• collector roadways
• residential streets
• parklands and campgrounds
Various guidelines of national and international lighting engineering and architectural groups may be
consulted for reasonable limits and strategies to optimize lighting designs limiting negative effects. The
following table shows some of these groups.
TABLE 2-2. Resources for Lighting Engineering Guidelines
Organization Internet Address
The International Commission on Illumination (CIE) http://www.cie.co.at/cie

The Illuminating Engineering Society of North America (IESNA) http://www.iesna.org
The European Committee for Standardization (CEN) http://www.cenorm.be
Energy Use
Important factors affecting energy use include light levels, lamp type, ballast type, and electrical sys-
tems quality. Lighting curfews can also help conserve energy (see Section 2.4 for more information).
Tourists, Businesses, and Nighttime Activities
Public lighting intended to attract tourists, CBD businesses, and nighttime activities may include
pedestrian, building façade, bridge or structure, landscape, and sign lighting. The intent is to provide a
pleasing and attractive environment for nighttime activities.
Planned Maintenance
The purpose of planned maintenance is to prioritize maintenance schedules, organize contracts, and
determine acceptable levels of service. Planned maintenance establishes consistent replacement products,
coordinates traffic control setups, enhances the use of electronic monitoring, takes advantage of modern
inventory systems, reduces trouble-shooting of problems, reduces electrical hazards, which can lower over-
all costs. Larger systems are more likely to have a lighting engineer available to oversee the total process
of design, installation, and maintenance, thereby allowing for replacement fixtures with suitable photo-
metric, electrical performance, and aesthetic features. Under a master lighting plan, life-cycle costs, life
expectancy, and quality studies are enhanced, and the region may have more buying clout. See Section
2.5 for a discussion of how modern electrical and lighting management systems can help with mainte-
nance.
2.4 LIGHTING CURFEWS
Introduction
Curfews for lighting involve the use of modern controls to turn off or dim selected parts of lighting
systems as permitted by reduced traffic flow, favorable weather conditions, and other local conditions.
Lighting curfews represent the active “operation” of the system, allowing for energy savings, greater
flexibility in resource allocation, and reduction of light-trespass. However, officials implementing such
6
options should be aware of consequences and conduct meaningful studies of costs and benefits.
Studies on the effectiveness of roadway lighting have mixed results. Generally, continuous lighting may
be expected to reduce nighttime crashes by about 30 percent. The precise mechanism for the reduction of
those crashes is not known. Currently it is not possible to translate surrogate measures, driver performance
with targets, or other measures directly into a safety change.
Motor vehicle crash data for 1998 from the National Highway Traffic Safety Administration’s Fatality
Analysis Reporting System (FARS) and the General Estimates System (GES) show that 90 percent of
fatal and injury crashes that occur on the roadway, where lighting guidelines specify that light be placed,
are multiple vehicle crashes. The number of overall crashes tapers off substantially after midnight on
weekdays and after 4:00 a.m. on weekends. At these late hours, most of the crashes are single vehicle, off-
roadway crashes for which lighting may not be likely to help, except possibly at decision-making points
such as ramp gores, intersections, and merge areas.
Warrants for lighting are empirically derived and based, among other things, on traffic volume. For light-
ing that has been installed based on traffic volumes, it may be reasonable to cut back the operation of the light-
ing system to complete interchange lighting or to partial interchange lighting when traffic volumes subside.
Studies show that crash rates increase where systems are turned off or where every other luminaire is
turned off. Alternate luminaire operation results in poor uniformity ratios. The issue of driver needs and safety
versus conservation efforts should be closely examined when considering curfews. Poorly conceived conser-
vation efforts may contribute to increases in traffic crashes and operational problems. These problems may
actually result in higher overall costs.
Reasons for Curfews
Government entities around the world are considering lighting curfews for the following reasons:
• Low late-hours traffic volumes. AASHTO warrants for highway lighting are based, among other
things, on traffic volume. When the primary reason that lighting is (or was) installed was due to high
traffic volumes and high usage, which drops off in later hours of the night, it is reasonable to turn off
or reduce the lighting after such drop-offs.
• Freeing resources for greater overall safety. Lowering the operational costs of lighting systems by
reducing electrical and maintenance costs through curfews may allow more lighting systems or other
crash countermeasures to be installed, thereby reducing the overall nighttime crash rate within the
jurisdiction of the master lighting plan.
• Technology now practical. Modern control technology now allows control of individual luminaires
or systems of luminaires at reasonable costs.
• Positive study results. Recent studies show that light dimming and turn-off curfews are viable
options for the management of public lighting systems, including roadway lighting. Although past
studies showed unfavorable increases in traffic crash rates because of turning off lighting, these stud-
ies were performed on systems where the lighting was turned off or partially turned off for the entire
nighttime period.
• Energy savings. Energy costs can be high.
• Sky glow issues. Sky glow issues are of increasing importance (see Chapter 12).
Considerations Before Implementation
Special events, maintenance or construction activities, weather, and other local considerations should
be included in the decision to implement lighting curfews. Implementation of curfews should occur
through traffic management centers or other central locations and should be monitored to gain experience
as to the best operational procedures.
The following excerpts are from the FHWA report Reduced Lighting During Periods of Low Traffic
Density (FHWA/RD-86/018). When considering lighting curfews, the complete report should be read,
7
while realizing that it is dated August 1985. Modern crash data should be analyzed.
In part, this report states, over 50 percent of all motor vehicle fatalities occur in darkness even
though only 25 percent of all travel occurs at night. This over representation has been used as a jus-
tification for installing fixed roadway lighting on many highways. However, research that has
attempted to determine the effect of such fixed lighting on frequency and severity of night accidents
appears to be mixed, such frequencies and severities being dependent on a host of geometric and
traffic factors including the volume of traffic utilizing the road, how such volume is related to the
road’s capacity, and the complexity of the driver’s visual search task.
During the past decade, several highway agencies have switched off roadway lighting during
periods of energy shortages to reduce maintenance and operating costs. However, quite often such
lighting was restored when nighttime accidents increased. One fundamental problem with these
light reduction techniques was that lighting was reduced or eliminated during the entire nighttime
period, rather than only when traffic volume was low.
By providing full lighting during periods when volumes are high and the roadway operates near
capacity and providing reduced lighting as the traffic decreases, the potential exists for realizing con-
siderable energy savings while still providing the benefits of full lighting at locations (e.g., inter-
changes) and at times (i.e., high volumes) where driver decision-making is the most critical and the
greatest visibility is required.
From a safety standpoint there is a definite reduction in (simulated) hazard detection perfor-
mance, which theoretically implies some reduction in safety. This implied reduction in safety is sta-
tistically significant for all off and one side only lighting tactics, but not statistically significant for
the dimmed tactics and the every other off tactic. Unfortunately, it is not possible at this time to quan-
tify the exact decrease in safety in terms of the frequency of nighttime accidents, the night accident
rate, or the night-to day accident ratio. Only an evaluation of long term installations can address this
issue (see following discussion of further research needs).
Reduced freeway lighting tactics normally should not be implemented before about 11:00 p.m.
in most urban areas, since traffic density typically remains relatively high until that time. Regularly
scheduled sports events and other large traffic generators could change this time to a later hour,
while cities with little or no evening activity might allow an earlier light reduction.
2.5 ELECTRICAL AND LIGHTING MANAGEMENT SYSTEMS
Introduction
Master lighting plans allow for benefits to be derived from modern electronic monitoring and control
systems known as electrical and lighting management systems (ELMS). This technology can detect light-
ing problems more quickly and accurately than traditional methods. The maximum benefit is most likely
realized when the technology is applied across the entire lighting infrastructure within a local authority’s
inventory. Remote monitoring technology can modernize the management, operation, and maintenance
of the street lighting infrastructure.
Cost
The costs of introducing the technology are difficult to estimate without significant evaluation. This
has led to concerns over initial investment costs, running costs, and anticipated payback on investment. It
is important that evaluations consider a whole system approach. Remote monitoring can be progressively
used and specified for new lighting schemes and for installations requiring additional or extraordinary
8
monitoring and control.
Benefits
The benefits of ELMS are outlined in the following table.
Benefits of ELMS
Benefit Explanation
More efficient contract management Contracts can be written to pay for maintenance by the
month instead of by each part repaired or replaced.
Percent of luminaires burning can be verified electroni-
cally through ELMS, thus reducing survey costs.
Improved electrical safety and reliability ELMS provides for ground fault detection and notifi-
cation of system malfunction.
Energy savings Energy saving lighting curfews are facilitated through
remote or automatic control of lighting systems.
Reduced maintenance costs ELMS facilitates advance knowledge of problems,
parts needed, and types of personnel needed. This
results in better overall contractor information and
supervision of maintenance work.
Reduced life-cycle costs ELMS can provide the information and documenta-
tion needed to write specifications and modify
approved products lists, thus lowering maintenance
costs and improving reliability.
The recent avalanche of technical advances is about to bring a wave of new technologies to organiza-
tions that operate and maintain street lighting systems. These technologies have the potential to improve
service quality, lower maintenance costs, increase productivity, and conserve energy. But the implemen-
tation of new technologies also holds risk because of technical and commercial complexities. The key to
reducing this risk is to make the new technologies compatible with existing systems, and to ensure that
the systems of different suppliers are interoperable.
Master lighting plans and ELMS allow coordination, through the traffic management centers, of all
interested subjects and parties including cities, department of transportation personnel, police, special
9
events, traffic management, state and local ordinances, landscape lighting, landscape installation (espe-
cially trees), landscape maintenance, pedestrians, cyclists, and central business district interests.
Benefits of ELMS Standardization
Interoperability and compatibility standards for ELMS results in four significant benefits, as outlined
in the following table.
Benefits of ELMS Standardization
Benefit Explanation
Reduction in the complexity of integrating This translates into lower risk of implementation
systems delays and fewer resources required for system instal-
lation and introduction.
Creation of more open and competitive market Proprietary technologies tend to lock buyers into sin-
gle-source supply arrangements. Such arrangements
give the incumbent supplier undue leverage in future
purchases and make substitution of non-performing
suppliers expensive. Standardization discourages this
tendency.
Allowance for the mixing of systems from Various systems have their own strengths and weak-
different suppliers nesses, given the differing situations in which they
are used. If a buyer wishes to implement a system
with “best of breed” components, then interoperation
and compatibility between these components make
that possible.
Evolution through open participation in which Due process through a public forum ensures validation
all views are considered and utility of the underlying technologies.
Interoperability and compatibility standards lead to a more competitive environment for suppliers.
However, suppliers benefit from standards as well. In general, buyers are suspicious of single-source
market conditions, and therefore simply delay purchases until a competitive environment has evolved.
History has shown that the establishment of standards accelerates adoption of new technologies and
causes the market to expand more rapidly.
2.6 CONDUCTING STUDIES
Introduction
Before developing a master lighting plan, a study should be conducted by traffic and lighting engi-
neers. The study is needed for the following reasons:
• to determine how the various lighting systems can best be optimized and coordinated
• to justify the implementation of lighting curfews
• to justify expenditures for technological improvements such as ELMS.
This section describes the recommended scope of such a study. Generally, the study should cover the
following items:
10
• electrical system
• purpose of the lighting system
• benefits and effects of curfews and dimming
• traffic studies
• surrounding land use and surrounding lighting
• security concerns
• sky glow and light trespass issues
• community goals
• coordination with navigation authorities
• traffic management interaction
• controls
• curfew monitoring
• ELMS Implementation
• coordination with roadway maintenance and construction activities
• budget.
These items are discussed in this chapter and following chapters under separate headings.
Electrical System
The study should evaluate electrical energy use and potential for savings.
Benefits and Effects of Curfews and Dimming
Dimming should be considered if light levels are higher than those required by AASHTO. In some
instances lighting is installed at levels higher than the minimum required because of safety or work zones.
It may be reasonable to dim this to only the required level after heavy traffic subsides. The same logic can
be applied to dimming as is applied to curfews.
Traffic Studies
Assess the hourly nighttime traffic volumes and crashes. Consider reducing continuous lighting to par-
tial interchange lighting at the point during the nighttime hours when fatal crashes, analyzed over a three-
year period, is found to drop to half of the peak rate. Consideration may be given to the type of crashes
and a study to determine if lighting is a factor in the specific traffic safety problem.
Community Goals
Conduct a survey of community goals for lighting.
Budget
Consider how the master lighting plan relates to the following budgetary factors:
• How will the plan affect the budget for lighting installation and maintenance?
• How will the plan affect the budget and methods for other traffic safety devices?
• What is the budget for installing controls?
11
Chapter 3
Techniques of Lighting Design
3.1 OVERVIEW
Introduction
Industry development and general experience on lighting of roadways has resulted in a reasonably
well-developed technique for the design of lighting systems. Accepted methods exist for achieving a
given lighting condition known as either level of illuminance or level of luminance. These methods per-
mit ready analysis of alternative lamps, luminaires, mounting height, luminaire spacing, energy con-
sumption, etc., to determine a preferred design. As more efficient light sources and better luminaires are
developed, they are incorporated into current designs.
The design of a roadway lighting installation is a process of applying known or specified photomet-
ric characteristics of selected lamp-luminaire combinations. Since illuminance considers only incident
light, photometric information about the luminaire and proximity and orientation to the pavement sur-
face is the only required information. To achieve the desired luminance, pavement reflectance should be
known. A trial-and-adjust process of assumed luminaire locations is used in making calculations of
either the average amount of illuminance or the average luminance over the roadway. For each lamp-
luminaire combination, the manufacturers’ data will show photometric performance of luminaires.
These data can be used, along with pavement reflectance characteristics in the case of luminance, to the-
oretically determine luminaire positions to produce the desired luminance over the given pavement area.
It is important to test luminaires to determine that luminaires supplied match the photometric perfor-
mance of luminaires used in the design process. Experience has shown that disregard for testing has
yielded disappointing results.
Calculations are performed by computer programs designed for lighting design purposes. The values
of the average amount of illuminance or luminance and variations in the average level are the basis of
design.
Level and Uniformity of Illuminance and Luminance
The level and uniformity of illuminance or luminance along a highway depends on several factors,
including the lumen output of the light source, luminaire distribution, mounting height, luminaire posi-
tion, pavement reflectance, and spacing and arrangement. The same average level can be obtained by dif-
ferent installation arrangements, such as a few high-output light sources or a greater number of
low-output sources. A concern in comparing of such alternate systems is the uniformity of illuminance or
luminance over the traveled way to be lighted.
Luminaires are available in a wide range of types, sizes, and photometric outputs. Lighting systems
using high-efficacy and large-lumen lamps can be designed to provide a satisfactory level and unifor-
mity of illuminance or luminance. Higher luminaire mounting heights offer a number of advantages and
should have full consideration when selecting design criteria. All appropriate light sources should be
considered, and the size or sizes that will give the most effective and economical lighting system should
be used.
13
Illuminance Considerations
Illuminance in roadway lighting is a measure of the light incident on the pavement surface. It is mea-
sured in foot-candles (Lux). Illuminance and luminance values are empirically derived from many years
of experience. Illuminance design may be accomplished with lighting design software or with iso-foot-
candle templates. The illuminance at any certain point will be the sum of illuminance from one or several
contributing sources.
Luminance Considerations
Luminance in roadway lighting is a measure of the reflected light from the pavement surface that is
visible to the motorist’s eye.
Pavement Reflectance Important. In addition to knowing the position, distribution, and intensity of
the luminaire, the calculation of luminance requires a knowledge of the reflectance characteristics of the
pavement and the location of the observer. See Fig. 3-1.
A system of pavement reflectance values used since 1970 by C.I.E. and adopted by ANSI/IESNA
divides the reflectance characteristics of pavements into four categories: R1, R2, R3, and R4. (Tables A3-
A6 CIE 132-1999 and Tables A1-A4 ANSI/IESNA RP-8-00). Each category has its own table of values
of reflectance for specified angles and are commonly known as r-tables. The r-tables have been adopted
for the purpose of this guide. Road surface classifications are described in Table 3-1. When designing a
roadway lighting system using these classifications, the final type of road surface in place, when the light-
ing system is operative is usually selected. (See the online Literature Review at http://downloads.trans-
portation.org/lighting.pdf for more information on pavement reflectance).
A calculation of reflected light toward the eye of the observer is made for each roadway point 83 m
from the observer, summing the luminance from each luminaire.
Figure 3-1. Calculation points for luminance and illuminance design methods.
14
TABLE 3-1. Road Surface Classifications
Class Qo* Description Mode of Reflectance
R1 0.10 Portland cement concrete road surface. Asphalt Mostly diffuse

road surface with a minimum of 12 percent of the
aggregates composed of artificial brightener
(e.g., Synopal) aggregates (e.g., labradorite, quartzite).
R2 0.07 Asphalt road surface with an aggregate composed Mixed (diffuse and
of minimum 60 percent gravel [size greater than specular)
1 cm (0.4 in.)].
Asphalt road surface with 10 to 15 percent artificial
brightener in aggregate mix. (Not normally used in
North America).
R3 0.07 Asphalt road surface (regular and carpet seal) with Slightly specular
dark aggregates (e.g., trap rock, blast furnace slag);
rough texture after some months of use (typical
highways).
R4 0.08 Asphalt road surface with very smooth texture. Mostly specular
* Qo 5 representative mean luminance coefficient.

Reprinted from American National Standard Practice for Roadway Lighting, ANSI/IES RP-8-00, Illuminating
Engineering Society of North America. Used by permission.
The observer moves with points. For example:

• The calculation for point Ap is made from observer position Ao.
• The calculation for point Bp is made from observer position Bo, same one quarter lane line.
• The calculation for point Cp is made from observer position Co, lane 4.
The number of points to be considered is calculated as follows:
Luminaire Cycle Distance

Number of Points =
10
The distance between points should not exceed 15 feet (5 m). Calculations should include a minimum
of three luminaire cycles downstream and one luminaire cycle upstream from reference (0.0) REF.
Luminance calculations place the observer’s (motorist’s) eye height at 4.8 ft. (1.45 m) above grade.
The 4.8 ft. (1.45 m) is a design figure used internationally and does not affect the driver eye height of
3.5 ft. (1.07 m). The observer’s line of sight is downward at one degree below horizontal and parallel to
the edge of the roadway along lines one-quarter roadway lane width from the edge of each lane. The
observer is positioned at a point 272 ft. (83.07 m) before the first point in the cycle to be evaluated.
Luminance Uniformity Calculation. Luminance uniformity is calculated both as the ratio of average
level to minimum point and maximum point to minimum point. The average-to-minimum-point method
uses the average luminance of the roadway design area between two adjacent luminaires divided by the
lowest value at any point in the area. The maximum-to-minimum-point method uses the maximum and
minimum values between the same adjacent luminaires. Luminance uniformities correlate closely with
the observer’s ability to detect differences in brightness levels.
15
Glare Evaluation. In conjunction with the luminance method, the evaluation of glare from the fixed
lighting system is relevant and included with the luminance criteria. The disability glare (veiling luminance)
has been quantified to give the designer the information to identify the veiling effect of glare as a percent
of average overall luminance.
Visibility-Based Design Methods
An alternate design method named small target visibility (STV) was proposed by the Illuminating
Engineering Society of North America (IESNA) in RP-8-1990 and was not adopted. With some mod-
ification, the STV method was adopted as RP-8-2000. The STV method has not been adopted by this
guide, because it does not adequately describe visibility in the roadway scene for the following
reasons:
• The research shows that minor changes in the roadway lighting system that may cause large changes
in the target visibility level (VL) values—because only the targets and their immediate background
are considered—may not have a significant effect on the information content of the overall scene.
Isolation on the VL of a small target is clearly too restrictive.
• The potential error in the calculation of target visibility is unacceptably high for engineering
calculations.
• Calculations do not include off-roadway sources or headlamps of multiple vehicles, as is typically
found in practice.
• The visibility of objects varies substantially with respect to observation distance. Static observation
distance does not describe the visibility of small targets in general on a roadway and certainly does
not describe the overall visibility of the roadway environment.
• The visibility of small targets on an empty roadway does not describe driver information needs suf-
ficiently to be used as a measure of safety.
• STV methods require too many simplifying assumptions that introduce unrecognized error into the
result. This makes it an approximation at best, and totally inaccurate at worst.
• The contribution of pavement reflectance to visibility-based design is not adequately recognized.
Background luminance is one of two fundamental design parameters. Before a reliable visibility-
based design methodology can be developed, modern pavement reflectance data must be
obtained, the change in pavement reflectance with respect to traffic must be well known, and the
effect of variable pavement reflectance on contrast must be known. In addition, the effect of dri-
ver observation angle change must also be fully understood before pavement reflectance can be
accurately estimated.
• The validity of the design calculations are further questioned because many of the physical para-
meters used are variable over time. The pavement’s reflective characteristics will change with
age. The luminaires will accumulate dirt and burn out, thus changing their output characteristics.
The amount of off-road lighting that contributes to visibility on the road changes as development
along the lighted area changes. Finally, normal weather variations such as rain and ice totally
invalidate the design calculations by changing the pavement’s reflective characteristics from dif-
fuse to specular.
Recent research on information theory (and the calculation tools provided by similarity theory) (1) fur-
nish a powerful and attractive tool for analyzing roadway lighting design. The combination of digital
imaging and IT-based processing algorithms was successfully used to quantify a predictable function of
visibility. These are likely to be the future of visibility based designs in lieu of target visibility.
16
3.2 WARRANTING CONDITIONS
General
This discussion summarizes available conclusions for the selection of those sections of freeways on
which fixed source lighting may be warranted. These warrants may be applied to roadways other than
freeways, as practicable.
Benefits of Lighting
The investment of public funds in roadway lighting returns benefits to the public in several ways.
Lighting benefits motorists by improving their ability to see roadway geometry and other vehicles at
extended distances ahead. This results in greater driver confidence and improved safety, particularly in
inclement weather. Lighting may also improve roadway capacity. Other benefits include improved pedes-
trian safety, improved public safety and security, convenience, and civic pride and recognition.
It is generally recognized that traffic safety benefits resulting from fixed source lighting correspond in
some proportion to traffic volume. High volume is inevitable on freeways in urban and suburban areas.
The number of stalled vehicles and other emergency stops is proportional to traffic volume.
Rural Interchanges
An unlighted rural interchange often presents unique conditions that require special consideration
before conclusions regarding lighting can be reached. Rural interchanges normally have ample space for
sign installation. Where the interchange type and detail are typical of most others on the freeway, and a
delineator system is included, the diverging and merging areas may be well understood and reasonably
discernible without lighting. However, installation of a few lighting units at the point of on- or off-move-
ments and ramp terminals could contribute to driver ease by providing visual indication for the maneuver
areas. Many rural interchanges with low traffic volumes do not warrant roadway lighting. However, there
are circumstances under which partial interchange lighting is appropriate and still other conditions where
complete interchange lighting is the preferred treatment.
Warranting Conditions
The warranting conditions in this section are for the purpose of establishing a basis on which lighting
for freeways may be justified. The warrants provide minimum conditions to be met whenever an agency
is contemplating lighting for new or existing facilities. Meeting of the warrants does not obligate the
highway agency to provide lighting. Jurisdictions may adopt higher numerical warranting values or more
restrictive subjective values for local use.
Master lighting plan ideas may be implemented in order to provide lighting at the times and places
most likely to provide the highest benefit to citizens.
The decision to provide lighting, and the extent of that lighting once warranting conditions are met,
rests with the highway agency. In determining when and where lighting is to be provided, the agency may
adopt a policy that includes these warrants and is also based on such factors as availability of funds, traf-
fic and crash data, roadway safety rating, etc. The warrants are not to be construed as the only criteria for
justifying lighting. The warrants are intended to be an easily understood tool to assist administrators and
designers in considering lighting on roadways.
Local conditions, such as frequent fog, ice, snow, roadway geometry, ambient lighting, sight distance,
signing, etc., could justify modification of these warrants.
17
The following are separate warrants for continuous freeway lighting, complete interchange lighting,
and partial interchange lighting.
Continuous Freeway Lighting
Description. A continuous lighting system provides relatively uniform lighting on all main lanes and
direct connections, and complete interchange lighting of all interchanges within the section. Frontage
roads are not normally continuously lighted. The lighting units may be conventional luminaires or high
mast assemblies or both.
Continuous lighting may be warranted under one of the conditions described in the following table.
TABLE 3-2. Warranting Conditions for Continuous Freeway Lighting (CFL)
Case Warranting Conditions
CFL-1 Sections in and near cities where the current average daily traffic (ADT) is 30,000 or greater.
CFL-2 Sections where three or more successive interchanges are located with an average spacing of
1.5 miles or less, and adjacent areas outside the right-of-way are substantially urban in
character.
CFL-3 Sections of two miles or more passing through a substantially developed suburban or urban
area in which one or more of the following conditions exist:
a. local traffic operates on a complete street grid having some form of street lighting, parts
of which are visible from the freeway
b. the freeway passes through a series of developments—such as residential, commercial,
industrial and civic areas, colleges, parks, terminals, etc. that include lighted roads,
streets, parking areas, yards, etc.—that are lighted
c. separate cross streets, both with and without connecting ramps, occur with an average
spacing of 0.5 miles or less, some of which are lighted as part of the local street system
d. the freeway cross section elements, such as median and borders, are substantially reduced
in width below desirable sections used in relatively open country.
CFL-4 Sections where the ratio of night to day crash rate is at least 2.0 times the statewide average for
all unlighted similar sections, and a study indicates that lighting may be expected to result in a
significant reduction in the night crash rate. Where crash data are not available, rate
comparison may be used as a general guideline for crash severity.
Definition. Complete interchange lighting is defined as a lighting system that provides relative uni-
form lighting within the limits of the interchange, including:
• main lanes
• direct connections
• ramp terminals
• frontage road or crossroad intersections.
18
Complete interchange lighting may be warranted under one of the conditions described in the follow-
ing table.
TABLE 3-3. Warranting Conditions for Complete Interchange Lighting (CIL)
CIL-1 Where the total current ADT ramp traffic entering and leaving the freeway within the
interchange areas exceeds 10,000 for urban conditions, 8,000 for suburban conditions, or
5,000 for rural conditions.
CIL-2 Where the current ADT on the crossroad exceeds 10,000 for urban conditions, 8,000 for
suburban conditions, or 5,000 for rural conditions.
CIL-3 Where existing substantial commercial or industrial development that is lighted during hours
of darkness is located in the immediate vicinity of the interchange, or where the crossroad
approach legs are lighted for 0.5 mile or more on each side of the interchange.
CIL-4 Where the ratio of night to day crash rate within the interchange area is at least 1.5 times the
statewide average for all unlighted similar sections, and a study indicates that lighting may be
expected to result in a significant reduction in the night crash rate.
Where crash data are not available, rate comparison may be used as a general guideline for
crash severity.
Partial Interchange Lighting
Definition. Partial interchange lighting is defined as a lighting system that provides illumination only
of decision making areas of roadways including:
• acceleration and deceleration lanes
• ramp terminals
• crossroads at frontage road or ramp intersections
• other areas of nighttime hazard.
Partial interchange lighting may be warranted under one of the conditions described in the following
table.
TABLE 3-4. Warranting Conditions for Partial Interchange Lighting (PIL)
PIL-1 Where the total current ADT ramp traffic entering and leaving the freeway within the
interchange area exceeds 5,000 for urban conditions, 3,000 for suburban conditions, or 1,000
for rural conditions.
PIL-2 Where the current ADT on the freeway through traffic lanes exceeds 25,000 for urban
conditions, 20,000 for suburban conditions, or 10,000 for rural conditions.
PIL-3 Where the ratio of night to day crash rate within the interchange area is at least 1.25 times the
statewide average for all unlighted similar sections, and a study indicates that lighting may be
expected to result in a significant reduction in the night crash rate.
Where crash data are not available, rate comparison may be used as a general guideline for
crash severity.
19
Special Considerations
Continuous, complete-interchange or partial-interchange lighting is considered justified where the

local governmental agency finds sufficient benefit in the forms of convenience, safety, policing, commu-
nity promotion, public relations, etc. to pay an appreciable percentage of the cost of or wholly finance the
installations, maintenance, and operation of the lighting facilities. For detailed discussions of these issues,
see the section on Master Lighting Plans.
Where there is continuous freeway lighting, there should be complete interchange lighting. When con-
tinuous freeway lighting is warranted, but not initially installed, partial interchange lighting is considered
justified under the continuous freeway lighting warrants CFL-1 or CFL-2. This would preclude the
requirements of satisfying the partial interchange lighting warrants PIL-1 or PIL-2.
Where complete interchange lighting is warranted, but not initially fully installed, a partial lighting
system that exceeds the normal partial installation in number of lighting units is considered justified.
Lighting of crossroad ramp terminals is warranted regardless of traffic volumes, where the design
requires the use of raised channelizing or divisional islands, or where there is poor sight distance.
Bridges
It may be desirable to provide fixed source lighting on long bridges in urban and suburban areas even
though the approaches are not lighted. On bridges without full shoulder, lighting enhances both safety and
utility of the bridges. Where bridges are provided with sidewalks for pedestrian movements, lighting is
warranted for pedestrian safety and policing.
3.3 DESIGN VALUES FOR FREEWAYS
General
The previously described warranting conditions should be used in deciding whether and where fixed
source lighting will be installed. Where continuous lighting is to be provided, lighting should be designed
to comply with the requirements of the Illuminance and Luminance Design Values Table (Table 3-5) for
either the Illuminance design method or the Luminance design method. Both methods should meet the
Veiling Luminance Ratio requirements.
Design Considerations
The selection of light source, luminaire distribution, mounting height and luminaire overhang is an
engineering decision based on such factors as geometry and character of the roadway, environment,
proposed maintenance, economics, aesthetics, and overall lighting objectives. Maintenance consider-
ations for roadway lighting design are covered in Chapter 11. Light loss factors are covered in the
online appendices located at http://downloads.transportation.org/lighting.pdf.
Lighting Levels on Crossroads
Lighting levels on the crossroad approaches should not be reduced through the interchange area. If
existing crossroad illuminance currently is deemed inadequate, it should be considered for upgrading.
20
Where partial interchange lighting is provided, luminaires should be located to best light the through
lanes and speed change lanes at diverging and merging locations (decision-making areas). Figure 3-2
shows examples of partial interchange lighting as used by some agencies with separate illustrations for
different ramp conditions. Other agencies may find more or fewer lighting units than shown in Figure 3-2
to be appropriate for their needs.
Figure 3-2. Typical Luminaire Locations—Partial Interchange Lighting.
21
Adaptation (Transition) Lighting
Rapid and extreme changes in levels of illuminance or luminance, especially those occurring when
leaving an area lit to levels much higher than normal roadway areas may be compensated for by using
transition lighting or adaptation techniques. The designer should recognize this phenomenon. Considera-
tion may be given to various methods to avoid rapid and extreme changes in light levels.
Bridges and Overpasses
On continuously lighted freeways and lighted interchanges, the lighting of bridges and overpasses
should be at the same level as the roadway. It is a practice by some agencies to consider the roadway
width to extend to the parapet, curb, or abutment, whenever the proximity of such barriers constitutes a
hazard.
Lighting poles on bridges should be located within the protection of railings or parapets. Poles located
on the inside of curves of superelevated bridges may be vulnerable to collision by trucks overhanging the
inside curb line.
Other Considerations
Aesthetics. The installed lighting system should have a pleasant daytime appearance. The design
should reflect aesthetic considerations.
Future Lighting. Provisions for present or future lighting may be included with roadway and struc-
tural work. Such provisions include underpavement conduit, concrete encased junction boxes and con-
duit, and pole anchorage in structures.
Area Classifications
Commercial. That portion of a municipality in a business development where ordinarily there are large
numbers of pedestrians and a heavy demand for parking space during periods of peak traffic or a sus-
tained high pedestrian volume and a continuously heavy demand for off-street parking space during busi-
ness hours. This definition applies to densely developed business areas outside of, as well as those that
are within, the central part of a municipality.
Intermediate. That portion of a municipality which is outside of a downtown area but generally within
the zone of influence of a business or industrial development, often characterized by a moderately heavy
nighttime pedestrian traffic and a somewhat lower parking turnover than is found in a commercial area.
This definition includes densely developed apartment areas, hospitals, public libraries, and neighborhood
recreational centers.
Residential. A residential development, or a mixture of residential and commercial establishments,

characterized by few pedestrians and a low parking demand or turnover at night. This definition includes
areas with single family homes, townhouses, and/or small apartments. Regional parks, cemeteries, and
vacant lands are also included.
22
3.4 STREETS AND HIGHWAYS OTHER THAN FREEWAYS

(INCLUDING WALKWAYS AND BICYCLE WAYS)
General
The literature is replete with data demonstrating the value of fixed lighting for facilities without access
control and the resulting benefits to the public. Some of the elements that warrant the lighting of urban
streets and highways are traffic volumes (both vehicles and pedestrians), at-grade intersections, turning
movements, signalization, and varying geometrics. The need for street and highway lighting in areas with
frequent inclement weather should be considered. In addition to its safety benefits, lighting may serve as a
crime deterrent, may aid law enforcement agencies, may contribute to user comfort, and often contributes
to community pride. These benefits may serve as a basis for the local government agency to pay an appre-
ciable percentage of the cost of, or wholly finance, the installation, maintenance, and operation of the
lighting facilities.
Warranting Conditions
It is not practical to establish specific warrants for the installation of roadway lighting that satisfy all
conditions. In general, lighting may be considered for those locations where the relevant governmental
agencies agree that lighting would contribute substantially to the safety, efficiency, and comfort of vehic-
ular or pedestrian traffic.
Lighting may be provided for all major arterials in urbanized areas and for locations or sections of
streets and highways where the ratio of night to day crash rates is higher than the statewide average for
similar locations, and a study indicates that lighting would significantly reduce the nighttime crash rate.
Rural Highways. Lighting of spot locations in rural areas should be considered whenever the driver is
required to pass through a section of road with complex geometry or raised channelization. The lighting
design treatment is typically similar to that for freeway ramp terminals.
Lighting Design Levels
The suggested lighting design values are provided in Table 3-5. Either the illuminance or luminance
technique may be used in the design of roadway lighting to meet the values set forth.
Average maintained luminance and illuminance levels in the tables are minimum levels when the out-
put of the lamp and luminaire is diminished by the maintenance factor.
Other Considerations
In using the values shown in Table 3-5, bear in mind that these values are for continuous lighting at
non-intersections. Special conditions may make somewhat different luminance and illuminance levels
desirable or necessary. For example, intersections with high pedestrian traffic volumes, curbs or divi-
sional islands may require somewhat higher levels. The lighting designer should use all available perti-
nent information in reaching a decision regarding the lighting level for any specific street or highway.
Intersections of two continuously lit streets are typically lit to a value equal to the sum of the individual
lighting level values.
There are many locations where very high levels of luminance or illuminance are provided for streets
in the central city business district. The reasoning is basically a commercial consideration and directed
toward making the downtown business area more appealing to visitors. Considerably higher luminance
or illuminance levels than those levels shown in the table may be justified on some basis other than
enhancing the safe and efficient flow of traffic.
23
24
TABLE 3-5a. Illuminance and Luminance Design Values (English)
Additional
Illuminance Method Luminance Method Values
(both Methods)
Roadway Off-Roadway
and Light Sources
Average Maintained Illuminance Average Maintained Luminance Veiling
Minimum Illuminance
Walkway Luminance
Illuminance Uniformity Ratio Ratio
Classification R1 R2 R3 R4 Lavg Uniformity
(foot-candles) (foot-candles) (foot-candles) (foot-candles) (foot-candles) avg/min cd/m2 Lavg/Lmin Lmax/Lmin Lv(max)/Lavg
General Land Use (min) (min) (min) (min) (max) (6) (min) (max) (max) (3)
(max)
Principal Arterials
Interstate and other freeways Commercial 0.6 to 1.1 0.6 to 1.1 0.6 to 1.1 0.6 to 1.1 0.2 3:1 or 4:1 0.4 to 1.0 3.5:1 6:1 0.3:1
Intermediate 0.6 to 0.9 0.6 to 0.9 0.6 to 0.9 0.6 to 0.9 0.2 3:1 or 4:1 0.4 to 0.8 3.5:1 6:1 0.3:1
Residential 0.6 to 0.8 0.6 to 0.8 0.6 to 0.8 0.6 to 0.8 0.2 3:1 or 4:1 0.4 to 0.6 3.5:1 6:1 0.3:1
Other Principal Arterials Commercial 1.1 1.6 1.6 1.4 3:1 1.2 3:1 5:1 0.3:1
(partial or no control of access) Intermediate 0.8 1.2 1.2 1.0 3:1 0.9 3:1 5:1 0.3:1
Residential 0.6 0.8 0.8 0.8 3:1 0.6 3.5:1 6:1 0.3:1
Minor Arterials Commercial 0.9 1.4 1.4 1.0 4:1 1.2 3:1 5:1 0.3:1
Intermediate 0.8 1.0 1.0 0.9 4:1 0.9 3:1 5:1 0.3:1
Residential 0.5 0.7 0.7 0.7 4:1 0.6 3.5:1 6:1 0.3:1
Collectors Commercial 0.8 1.1 1.1 0.9 4:1 0.8 3:1 5:1 0.4:1
Intermediate 0.6 0.8 0.8 0.8 4:1 0.6 3.5:1 6:1 0.4:1
Residential 0.4 0.6 0.6 0.5 4:1 0.4 4:1 8:1 0.4:1
Local Commercial 0.6 0.8 0.8 0.8 6:1 0.6 6:1 10:1 0.4:1
Intermediate 0.5 0.7 0.7 0.6 6:1 0.5 6:1 10:1 0.4:1
Residential 0.3 0.4 0.4 0.4 6:1 0.3 6:1 10:1 0.4:1
Alleys Commercial 0.4 0.6 0.6 0.5 6:1 0.4 6:1 10:1 0.4:1
As uniformity ratio allows
Intermediate 0.3 0.4 0.4 0.4 6:1 0.3 6:1 10:1 0.4:1
Residential 0.2 0.3 0.3 0.3 6:1 0.2 6:1 10:1 0.4:1

Sidewalks Commercial 0.9 1.3 1.3 1.2 3:1
Intermediate 0.6 0.8 0.8 0.8 4:1
Residential Use illuminance requirements
0.3 0.4 0.4 0.4 6:1

(2) All 1.4 2.0 2.0 1.8 3:1
Pedestrian Ways and Bicycle Ways
Notes: 1. Meet either the Illuminance design method requirements or the Luminance design method requirements and meet veiling luminance requirements for both the Illuminance and the Luminance design methods.
2. Assumes a separate facility. For Pedestrian Ways and Bicycle Ways adjacent to roadway, use roadway design values. Use R3 requirements for walkway/bikeway surface materials other than the pavement types shown.
Other design guidelines such as IESNA or CIE may be used for pedestrian ways and bikeways when deemed appropriate.
3. Lv(max) refers to the maximum point along the pavement, not the maximum in lamp life. The Maintenance Factor applies to both the Lv term and the Lavg term.
4. There may be situations when a higher level of illuminance is justified. The higher values for freeways may be justified when deemed advantageous by the agency to mitigate off-roadway sources.
5. Physical roadway conditions may require adjustment of spacing determined from the base levels of illuminance indicated above.
6. Higher uniformity ratios are acceptable for elevated ramps near high-mast poles.
7. See AASHTO publication entitled, "A Policy on Geometric Design of Highways and Streets" for roadway and walkway classifications.
TABLE 3-5b. Illuminance and Luminance Design Values (Metric)
Additional
Illuminance Method Luminance Method Values
(both Methods)
Roadway Off-Roadway
and Light Sources
Average Maintained Illuminance Minimum Illuminance Average Maintained Luminance Veiling
Walkway Luminance
Illuminance Uniformity Ratio Ratio
Classification R1 R2 R3 R4 Lavg Uniformity
(Lux) (Lux) (Lux) (Lux) (Lux) avg/min cd/m2 Lavg/Lmin Lmax/Lmin Lv(max)/Lavg
General Land Use (min) (min) (min) (min) (max) (6) (min) (max) (max) (3)
(max)
Principal Arterials
Interstate and other freeways Commercial 6 to 12 6 to 12 6 to 12 6 to 12 2 3:1 or 4:1 0.4 to 1.0 3.5:1 6:1 0.3:1
Intermediate 6 to 10 6 to 10 6 to 10 6 to 10 2 3:1 or 4:1 0.4 to 0.8 3.5:1 6:1 0.3:1
Residential 6 to 8 6 to 8 6 to 8 6 to 8 2 3:1 or 4:1 0.4 to 0.6 3.5:1 6:1 0.3:1
Other Principal Arterials Commercial 12 17 17 15 3:1 1.2 3:1 5:1 0.3:1

(partial or no control of access) Intermediate 9 13 13 11 3:1 0.9 3:1 5:1 0.3:1
Residential 6 9 9 8 3:1 0.6 3.5:1 6:1 0.3:1
Minor Arterials Commercial 10 15 15 11 4:1 1.2 3:1 5:1 0.3:1

Intermediate 8 11 11 10 4:1 0.9 3:1 5:1 0.3:1
Residential 5 7 7 7 4:1 0.6 3.5:1 6:1 0.3:1
Collectors Commercial 8 12 12 10 4:1 0.8 3:1 5:1 0.4:1

Intermediate 6 9 9 8 4:1 0.6 3.5:1 6:1 0.4:1
Residential 4 6 6 5 4:1 0.4 4:1 8:1 0.4:1
Local Commercial 6 9 9 8 6:1 0.6 6:1 10:1 0.4:1

Intermediate 5 7 7 6 6:1 0.5 6:1 10:1 0.4:1
Residential 3 4 4 4 6:1 0.3 6:1 10:1 0.4:1
Alleys Commercial 4 6 6 5 6:1 0.4 6:1 10:1 0.4:1

As uniformity ratio allows
Intermediate 3 4 4 4 6:1 0.3 6:1 10:1 0.4:1

Residential 2 3 3 3 6:1 0.2 6:1 10:1 0.4:1
Sidewalks Commercial 10 14 14 13 3:1

Intermediate 6 9 9 8 4:1
Residential
Use illuminance requirements
3 4 4 4 6:1

(2) All 15 22 22 19 3:1
Pedestrian Ways and Bicycle Ways
Notes: 1. Meet either the Illuminance design method requirements or the Luminance design method requirements and meet veiling luminance requirements for both the Illuminance and the Luminance design methods.
2. Assumes a separate facility. For Pedestrian Ways and Bicycle Ways adjacent to roadway, use roadway design values. Use R3 requirements for walkway/bikeway surface materials other than the pavement types shown.

Other design guidelines such as IESNA or CIE may be used for pedestrian ways and bikeways when deemed appropriate.
3. Lv(max) refers to the maximum point along the pavement, not the maximum in lamp life. The Maintenance Factor applies to both the Lv term and the Lavg term.
4. There may be situations when a higher level of illuminance is justified. The higher values for freeways may be justified when deemed advantageous by the agency to mitigate off-roadway sources.
5. Physical roadway conditions may require adjustment of spacing determined from the base levels of illuminance indicated above.
6. Higher uniformity ratios are acceptable for elevated ramps near high-mast poles.
7. See AASHTO publication entitled, "A Policy on Geometric Design of Highways and Streets" for roadway and walkway classifications.
25
3.5 POLE PLACEMENT GUIDELINES
Introduction
The careful and strategic placement of lighting poles along the roadway is an important concern for the
lighting engineer. Along with providing a lighting system that supplies the required levels of illumination,
it is important to provide lighting support structures are properly designed and carefully located to mini-
mize adverse effects on the traveling public.
The guidelines listed in this section have been established to simplify the pole placement process for
the lighting engineer. These guidelines address some of the more common situations encountered during
the design of a roadway lighting system.
Obstruction of View
Structural supports for lighting units should be designed and located so that they do not distract the
attention of the motorist or interfere with their view of the roadway and other important roadway features.
Supports should be placed so that they do not obstruct the view of signs.
Height Restrictions
The Federal Aviation Administration may have certain height restrictions for lighting poles placed
adjacent to airports and in their landing zones.
Medians
Locating structural supports for lighting units within a median area may be appropriate if the width of
the median is sufficient or if concrete median barriers are used. Locating poles in medians provides sev-
eral lighting and economic advantages that cannot be overlooked. Advantages include:
• The number of lighting poles is approximately half that required for house-side lighting.
• The amount of wiring required is reduced.
• Lighting otherwise wasted on the house side is used instead on the opposite roadway.
• Construction and maintenance costs are reduced.
• Visibility on the high-speed lanes is improved.
Disadvantages include:
• Inside lanes may need to be closed for maintenance of luminaires.
Gore Areas
Locating structural supports for lighting units within the clear zone of both a main lane and a ramp at
a gore area is not usually desirable unless located behind or atop a longitudinal traffic barrier or behind a
crash cushion.
Maintenance Considerations
The maintenance and servicing of roadway and sign lighting units should be considered when design-
ing the lighting system. Inspection for structural supports for lighting systems may detect corrosion or
fatigue problems. Some breakaway connections may require periodic maintenance so that the specified
26
torque requirements of the bolts are maintained so that the connection will function properly. The selec-
tion of the lighting pole location should also include consideration of the potential hazards imposed on
maintenance personnel servicing and maintaining the lighting equipment.
Adjacent to Deflecting Barriers
Structural supports for lighting units should not be located on the traffic side of guide rail or deflecting
barrier. In locating lighting poles behind rail, consideration should also be given to the distance necessary
for rail deflection in the barrier.
General Safety Concerns
On freeways, roadside design encompasses the area from lane edge line to lane edge line, in medians
and side separations (between main lanes and ramps or frontage roads) and from frontage road edge line
to right-of-way line. Roadside design on highways encompasses the area between the outside lane edge
and the right-of-way limits. A roadside environment free of fixed objects and with stable, flattened slopes
helps reduce crash severity when errant vehicles leave the roadway. The “forgiving roadside” concept
(AASHTO Roadside Design Guide) allows for the possibility of errant vehicles leaving the roadway, and
supports a roadside design that reduces the serious consequences of such an incident.
How Poles Fall on Impact. As a general rule, a luminaire support will fall near the line of path of a
vehicle after impact. Research shows (4) that 95 percent of “run off the road” vehicles depart the roadway
at an angle of 20 degrees or less. The mast arm on single mast arm poles usually rotates so that it is point-
ing away from the roadway when resting on the ground. This action generally prevents the pole from
going into other traffic lanes if sufficient “falling area” is provided. The provision of a “falling area” is
most beneficial on higher speed roadways. Sufficient falling area is generally considered to be 2/5 of the
pole mounting height.
Design Options
While the first three options (shown below) are the preferred choices, these solutions are not always
viable for the lighting engineer, because the lighting supports must remain near the edge of the roadway
in order to provide adequate levels of illumination on the roadway surface.
Option 1: Remove the Obstacle
Removing the obstacle may not always be a viable solution for the lighting engineer, because lighting
supports must be installed adjacent to the traveled roadway in order to provide adequate lighting levels on
the roadway surface. However, the number of lighting supports installed can be minimized by careful exam-
ination of the surroundings. For example, it may be possible to use a single support to serve more than one
purpose. Combination poles, serving multiple functions for lighting, traffic control, and electrical power,
should be considered as a way to reduce the number of poles along the edge of the traveled roadway.
Option 2: Redesign the Obstacle
Investigate options for redesigning the lighting system so that the lighting supports are located outside
the clear zone. This can be accomplished through higher mounting heights or off-set luminaires. The ulti-
mate design in this respect features the use of tower or high-mast lighting, because far fewer supports are
needed and they are located much farther away from the roadway. High-mast poles are an especially use-
ful method for lighting major interchanges.
27
Individual agencies must weigh the relative benefits of using off-set luminaires mounted on poles out-
side the clear zone against the benefits of using fixtures on break-away poles inside the clear zone which
can provide benefits of less veiling glare, lower light trespass and sky glow, and less visual clutter
Option 3: Relocate the Obstacle
Non-breakaway lighting poles should not be installed within the clear zone along the edge of the trav-
eled roadway unless protected from impact by guard rail or other barriers. The clear zone is defined as the
total unobstructed roadside border area desired. It starts at the edge of the traveled way and may include
shoulders, recoverable and non-recoverable slopes, and clear run-out areas. The clear zone is available for
the safe, unobstructed use and recovery of errant vehicles: placing non-breakaway lighting poles within
the clear zone should be avoided. Various agencies may have more or less extensive definitions and appli-
cations of clear zones. See the AASHTO Roadside Design Guide for more information on clear zones.
Option 4: Use Appropriate Breakaway Device
The design of the lighting system should include breakaway supports when the lighting standards can-
not be placed outside of the roadside clear zone or behind a longitudinal traffic barrier or crash cushion.
A breakaway support is a design feature that allows a luminaire support to yield, fracture, or separate near
ground level upon impact by an errant vehicle. See the AASHTO publication, Structural Supports for
Highways Signs, Luminaires, and Traffic Signals, for breakaway structural requirements.
When breakaway poles are placed inside the clear zone of freeways, they should be located at a mini-
mum of 15 feet (4.5 m) preferably 20 feet (6 m) from lane edge lines.
Breakaway devices should not be used outside the parameters that have been tested and approved by
FHWA. Mounting height and weight limits are particularly addressed by FHWA.
Angle of Impact. The breakaway mechanism is designed to activate when loaded in shear rather than
bending, and is designed to release in shear when impacted by a vehicle bumper at normal bumper height.
Locating supports along the edge of the roadway where they are likely to be loaded in bending rather than
in shear may result in more severe impacts and injuries to the occupants of the vehicle. Super-elevation,
side slope, rounding and offset, and vehicle departure angle and speed are design components that influ-
ence the striking height of a typical errant vehicle. If the breakaway support is struck at a point higher than
28 inches typical bumper height, the bending moment in the breakaway support may be sufficient to bind
the mechanism, resulting in non-activation of the breakaway device. For this reason, it is critical that the
breakaway supports are not located near ditches, on steep slopes, or at similar locations where a vehicle
is likely to be partially airborne at the time of impact. Limiting the negative side slopes to 1:6 between
roadway and luminaire supports should ensure acceptable striking height.
Surroundings. Breakaway supports should not be used in very confined urban areas with high pedes-
trian activity where there is a high possibility of a support striking a pedestrian, private property, or
another vehicle after impact.
Soil and Terrain. The type of soil surrounding a luminaire foundation may affect the performance of
the breakaway mechanism. If a support foundation is allowed to push through the soil, the luminaire sup-
port may bend rather than shear, resulting in the breakaway mechanism not being activated. After instal-
lation of the foundation, the surrounding terrain should be graded to permit vehicles to pass over any
non-breakaway portion of the support that remains in the ground or that is rigidly attached to the founda-
tion after impact.
Electrical Disconnects. The electrical circuitry contained within the breakaway support shall be
equipped with electrical disconnects to facilitate the breakaway mechanisms and to reduce the risk of
electrical shock from the exposed wiring after impact. The electrical disconnects must be used in all new
28
installations and considered for existing installations that experience frequent knockdowns. See Chapter
8 for more information on electrical requirements.
Conclusion. As a general rule, breakaway supports should be considered in some urban and most rural
areas, wherever vehicle speeds are moderate to high, or when the lighting standards cannot be placed
behind a longitudinal traffic barrier or crash cushion. The designer should weigh the relative risks
involved in these situations before selecting an appropriate design. If the use of breakaway supports is not
feasible, then shielding the obstacle may be the only viable solution.
Option 5: Shield the Obstacle
Only when the use of breakaway supports is not practical should a traffic barrier or crash cushion be
used exclusively to shield light poles.
Structural supports for high mast lighting systems should be placed outside the clear zone or they
should be protected with proper guide rail or other deflecting barrier. High mast lighting supports are con-
sidered fixed-base support systems that do not yield or break away on impact. The large mass of these
support systems and the potential safety consequences of the systems when they fall to the ground neces-
sitate a fixed-based design.
When a luminaire support is located near a traffic barrier, a breakaway bases may or may not be applic-
able, depending on the type and characteristics of the barrier. In general, the support should not be located
within the design deflection distance of the barrier or the railing should be strengthened locally to mini-
mize the resultant deflection distance.
Another alternative would be to mount the lighting standards on top of concrete median barriers, a
practice that oftentimes requires modifications to the lighting standard, median barrier, or both. For high
angle impacts into the concrete barrier or crashes involving trucks or buses, a luminaire mounted on top
of a concrete barrier may be struck. This type of installation generally does not use breakaway supports,
because of the risk a downed pole might present to opposing traffic or traffic below an elevated structure.
Poles should not typically be placed atop bridge rails on main lanes, connectors, or ramps that cross
other roadways since even non-breakaway poles are sometimes dislodged. Rather, the pole should be
placed on footings behind rails in such locations.
Option 6: Delineate the Obstacle
If the above alternatives are not appropriate or if a particular situation requires that additional measures
be taken, the designer should investigate delineating the obstacle to provide quicker recognition and
opportunity for response by the motorist. For example, providing reflectorized banding at the eye level of
the motorist for the poles that are in gore areas may provide quicker recognition by the motorist.
3.6 REFERENCES
1. AASHTO, A Policy on Geometric Design of Highways and Streets, 5th Edition, 2004.
2. AASHTO, Roadside Design Guide, 2002.
3. AASHTO, Standard Specifications for Structural Supports for Highway Signs, Luminaires, and Traffic
Signals, 4th Edition, 2001.
4. Rowan, N.J., Walton, N.E., (1972) Roadway Illumination Systems, Texas Department of Trans-
portation Research Report 137-2(F).
29
Chapter 4
High-Mast Lighting
High-mast lighting is comprised of groups of luminaires mounted on free-standing poles at mounting

heights that can vary from approximately 60 feet (18 meters) to 180 feet (55 meters). High mast poles are
usually provided with luminaire lowering devices that lower the luminaires for maintenance.
High-mast lighting is used principally at interchanges, toll plazas, rest areas, parking areas, and for
continuous lighting on highways that have wide cross sections.
Recommended lighting levels are on Table 3-5 in Chapter 3. Higher lighting levels than indicated in
Table 3-5 may be required after consideration of such factors as the complexity of the interchange, the
existence of high brightness from competing light sources near the roadway, and the prevailing level of
lighting on connecting roadways.
The benefits of high-mast lighting include excellent uniformity, lower glare, and fewer pole locations
when compared to conventional lighting systems. Traffic control requirements are reduced for mainte-
nance because the poles can be located out of the clear zone or recovery area and away from the roadway.
Surrounding off road areas receive incidental illumination that provides the motorist with a panoramic
view, compared to the “tunnel of light” effect provided by conventional systems. Another benefit is the
visibility of vertical surfaces of the roadway system such as guardrail, bridge columns, abutments and
drainage headwalls. High-mast lighting systems perform well under adverse weather conditions such as
rain, fog, and snow.
The most common type of luminaire used in high-mast lighting is the area type which is usually
offered with symmetric or asymmetric distribution. Both types of distribution are frequently used. Cutoff
style luminaires are recommended.
Scheduled inspections of the pole and lowering system may detect corrosion, fatigue, or other problems.
31
Chapter 5
Tunnels and Underpasses
5.1 OVERVIEW
The designer may review and become familiar with resources that exist related to the lighting design
of underpasses and tunnels. Literature is available on the technical aspects of visibility and lighting of
vehicular tunnels. A review of the research material included in the references of this guide as well as
other research efforts is encouraged and will provide considerably more detailed information than is pos-
sible in this guide. In addition to independent studies, IESNA and CIE offer recommendations for various
tunnel and underpass lighting applications.
Lighting levels are typically presented in luminance format. Surface reflectance, luminaire placement,
and luminaire orientation affect results. It is difficult to equate illuminance levels to luminance levels
given the various surface reflectance and luminaire orientation possibilities.
Future surface reflectance changes should be avoided unless the design allows for specific changes.
5.2 UNDERPASSES
General Scope and Guide Application
An underpass is defined as a portion of a roadway that extends through and beneath some natural or
man-made structure, which, because of its limited length to height ratio, requires no supplementary day-
time lighting. Underpasses of multiple highway structures will normally be treated separately where the
space between these structures permit good penetration of daylight on the underpass roadways, rather
than as one composite length. Daytime lighting may not be necessary in cases where the overhead struc-
ture allows relatively direct daylight penetration.
The specific geometry and roadway conditions, including vehicular and pedestrian activity, should
be considered in evaluating the need for daytime lighting. These features are more fully covered in the
tunnel lighting section.
Although tunnel designs are typically designed using luminance methods, underpass lighting can be
designed using the same method used for the roadway lighting design.
Warrants for Nighttime Underpass Lighting
Underpass lighting is warranted in areas that have frequent nighttime pedestrian traffic; or where
unusual or critical roadway geometry occurs adjacent to or in the underpass area.
Continuous lighting on the associated freeway lanes warrants the installation of underpass lighting.
Luminaires positioned adjacent to the underpass can often provide adequate lighting without supple-
mental underpass lighting.
33
Design Values for Underpass Lighting
Nighttime lighting levels and uniformities should target the lighting levels on the adjacent road-
ways. Higher levels of lighting may result because of luminaire mounting height and spacing limita-
tions. Increased levels should not exceed approximately twice that of the roadway adjacent to the
underpass.
High nighttime ambient brightness produced by lighting from other nearby sources may justify higher
lighting levels.
Selection and Placement of Underpass Luminaires
Luminaires attached to the structure along the roadside in full or partial view of the motorist may
necessitate glare control or the use of lower wattages. It is generally better to minimize source glare by
using several lower output luminaires than to provide one or two high output luminaires. The use of lower
lumen output fixtures tends to improve the uniformity of lighting while maintaining lighting levels.
Wall mount luminaires are usually easier to maintain, and they are less affected by structure vibration.
5.3 VEHICULAR TUNNELS
General Scope and Guide Application
A structure of any type that surrounds a vehicular roadway and is longer than an underpass is recog-
nized by this publication as a tunnel. Tunnels normally require supplementary day lighting to provide
adequate roadway visibility necessary for safe and efficient traffic operation.
Literature is available on the technical aspects of visibility and lighting of vehicular tunnels. A review
of the referenced material included in this guide as well as other research efforts is encouraged and will
Figure 5-1. Example of a Vehicular Tunnel.
34
provide considerably more detailed information than is possible in this guide.
Short Tunnels
A tunnel is considered “short” if its length from portal to portal is equal to or less than the wet
pavement minimum stopping sight distance as recommended by the latest AASHTO Policy on
Geometric Design of Highways and Streets for the vehicle operating speeds of the tunnel roadway and
approaches. A tunnel zone is a length of tunnel roadway equal to the wet pavement minimum stopping
sight distance. A short tunnel has only one zone.
Long Tunnels
A tunnel is considered “long” if its portal to portal length is greater than the wet pavement minimum
stopping sight distance. A long tunnel has two or more zones.
Warrants for Tunnel Lighting
The installation of daytime lighting is warranted when tunnel user visibility requirements are not met
with sunlight. Tunnel visibility factors include such items as the geometry of the tunnel and its
approaches, the traffic characteristics, the treatment of roadway and environmental reflective surfaces,
the climate, and the orientation of the tunnel.
The following criterion adapted from a procedure used in The Netherlands may be used as a guide for
deciding when daytime lighting should be provided. Refer to Figure 5-2. When viewing the tunnel from
the perspective of an approaching motorist and while the entire entrance is in view, if
1. The exit portal encompasses 50 percent or more of the entrance portal, daytime lighting is not likely
warranted.
2. The exit portal encompasses less that 20 percent of the entrance portal, daytime lighting is warranted.
3. If the condition is between 1 and 2 above, a detailed study should be conducted to determine if day-
time lighting is warranted.
Figure 5-2. Entrance and Exit Portal comparison method.
35
From Figure 5-2, the percent that the view of the exit portal occupies the view of entrance portal is:
100 * (EFGH)/(ABCD)
1. If rectangle EFGH is larger than 50 percent of rectangle ABCD then daytime tunnel lighting is not
recommended.
2. If rectangle EFGH is smaller than 20 percent of rectangle ABCD then daytime tunnel lighting is
recommended.
3. If the size of rectangle EFGH is between 20 percent and 50 percent of the size of rectangle ABCD
then a more detailed study is needed in order to formulate a recommendation.
Visibility Optimization of the Tunnel and Approach Features
It is important in the physical design of a tunnel structure that due consideration be given to lighting
needs.
The physical features of a tunnel can have a significant effect on reducing the day lighting needs.
The following items contribute to improved tunnel visibility and should be explored in the develop-
ment of daytime tunnel lighting designs.
1. Reduction of Ambient Daytime Brightness
Tunnel portals, adjacent walls, approach pavement, and other external features in the motorist’s
field of view should be darkened to the extent possible. Admixtures, overlays, vegetation, or other
methods that result in low reflectance, non-specular surfaces are recommended. Dark features
increase the degree of advance eye adaptation of the entering motorist and improves contrast with the
lower luminance levels in the tunnel interior. Tunnels with a predominant sky background above
the entrance should be reviewed for the use of plantings, screens, or panels that increase the size of
the darkened area above the portals.
2. Portal Design Factors
Upsweep ceilings may increase daylight penetration, but can result in increased tunnel structure
costs.
Sun screens have not been effective. Dirt accumulation, permanent depreciation of reflective
and light-transmitting properties, and snow and ice accumulation have posed serious problems. The
high initial costs of sun screens coupled with high maintenance costs have practically eliminated
their use.
3. Visibility Optimization of Tunnel Interiors
It is recommended that ceiling and wall surfaces be of an easily maintained finish with a nonspec-
ular reflective efficiency of at least 70 percent. High wall brightness is of great value in meeting vis-
ibility needs in tunnels that have curved roadways or approach roadways. Relatively narrow tunnels
where the width-to-height ratios are approximately 3 or less will develop interreflectivity that can
enhance tunnel visibility as a result of the reflected light from the walls.
Natural sunlight penetration in entrance portal areas can be improved by the use of wall, ceiling,
and roadway surface texture control. The use of vertical wall corrugations, coarse finished pave-
ments, or other treatments which produce surface relief, increase the retro-reflection of light enter-
ing the portal over that of smooth surfaces.
4. Types of Pavement Surfaces
The use of bituminous concrete on the approach road surface to the tunnel portal and portland
36
cement concrete on the road surface inside the portal for a distance at least equal to the safe sight
stopping distance will reduce the luminance contrast between the outside and the inside of the tun-
nel. This will in turn reduce entrance zone luminance and illuminance requirements. Future resur-
facing should account for the designed roadway surface.
5.4 DAYTIME LIGHTING OF TUNNEL INTERIORS
Short Tunnels—Silhouette Visibility
Short vehicular tunnels that have relatively straight and level approach alignments with corresponding
straight and level tunnel roadways may offer adequate visibility to the entering motorist by silhouette
viewing of other vehicles and objects on the roadway against the far side exit portal. These tunnels are
treated as underpasses in the guide. Silhouette visibility should be carefully evaluated with respect to the
tunnel geometry to provide visibility of objects within the tunnel. The roadway surface details will nor-
mally be indiscernible to the motorist with silhouette visibility.
In multi-lane one-way tunnels, or unseparated two-way tunnels, lighting should be provided to the
extent the motorist can distinguish lane markings or other delineation important to safe travel through the
tunnel.
Entrance Portal Lighting
The most critical portion of a tunnel that affects visibility is at the portal. This is commonly called the
“black hole” effect. Visibility of this first entrance zone, while still outside the tunnel, is essential to the
motorist in identifying and safely reacting to the presence of vehicles and objects that may be present on
the tunnel roadways. This is accomplished by lighting the entrance zone in proper proportion to the out-
side ambient luminance to which the motorists’ eyes are adapted.
The luminance of the approach pavement, adjacent landscape, sky, and the portal area itself, are all
integrated over time by the motorist’s eyes in adapting to overall ambient conditions. It is suggested that
an evaluation of brightness conditions be made for the actual roadway and tunnel prior to establishing a
lighting design.
A model simulation may be necessary for new facilities in order to duplicate the anticipated tunnel
approach conditions. The motorist’s field of view of adapted luminance should be evaluated at a location
along the approach roadway equal to the minimum stopping sight distance in advance of the portal. See
the CIE or IESNA tunnel lighting design guides for extensive discussion of these details.
Two and three lane one-way tunnels having favorable alignments of the approaches and tunnel struc-
ture, and which are of relatively short length, have been adequately lighted with relatively low artificial
lighting levels. The optimization of portal entrance conditions, in some cases, has produced adequate
entrance visibility at artificial luminance levels in the range of about 100 to 200 candelas per square meter
reflected from an in-service roadway surface.
Entrance zone lighting levels should be designed to accommodate the greatest ambient luminance
expected at the location. The stopping sight distance defined previously determines the length of entrance
zone lighting. Most tunnel approach roadways, except for extreme cases of vertical and horizontal curva-
ture, have entrance characteristics such that a point relatively close to the tunnel portal will confine the
motorist’s view to the predominance of the darkened tunnel structure. It is an acceptable practice to
include this “fixation” distance in the minimum stopping sight distance to reduce the length of the
entrance interior lighting. Preadaption should not normally be used to reduce portal lighting levels.
Lighting Beyond the Entrance Zone
If the tunnel is classified as a short tunnel, the entrance zone lighting level applies throughout its entire
37
length. However, in long tunnels, lighting beyond the minimum stopping sight distance should be
reduced progressively until an established minimum level is reached. It is recommended that beginning
at the end of the entrance zone lighting the levels be reduced in steps to a level not less than 5 horizontal
footcandles (54 lux) or 5 candelas per square meter on the roadways. Each stepped zone should have a
length at least equal to the minimum stopping sight distance.
Nighttime Tunnel Lighting
Nighttime lighting should, if practical, make use of a portion of the daytime lighting system, rather
than be a separate system. Nighttime levels in a tunnel should be somewhat higher, but not exceeding
three times that of the lighting requirements for the roadways adjacent to the tunnel. Uniformity of light-
ing should closely match that of the requirements for the adjacent roadways.
Tunnels located on non-continuously lighted roadways should be lighted to the minimum standards
required for the highway type and character as contained in this guide.
Selection and Placement of Tunnel Luminaires
The choice of particular types of tunnel luminaires and light sources should be made by considering
such items as luminous efficacy, source glare, light distribution characteristics, physical placement limi-
tations, frequency of maintenance, and resistance to damage. An important consideration in the choice of
a particular system for both daytime and nighttime lighting in long tunnels is the stroboscopic effect of
alternate bright-dark areas where luminaires do not provide a continuous line of luminance. Frequencies
in the range of 5 to 10 cycles per second have been observed to result in eye annoyance and should be
avoided at the particular design speed of the tunnel.
Tunnel Lighting Control Systems
Lighting levels for the entrance zone may be adjusted to match the ambient conditions created due to
varying light levels from season to season and during cloudy or inclement weather. If such system vari-
ances are determined to be economical and feasible, lighting levels in subsequent tunnel zones should
vary in the same proportion. Lighting systems for tunnels should be designed as fail safe as practical to
reduce the possibility of a total tunnel outage in the event of a circuit failure or other malfunction.
Maintenance Factor for Tunnel Lighting Design
The reduction of initial lighting levels becomes an important factor in tunnel lighting design. Initial
design levels should consider the frequency and degree of maintenance that is to be performed. Factors in
the range of 50 percent are commonly applied to tunnel lighting designs.
38
Chapter 6
Work Zone Lighting and Temporary
Roadway Lighting
Introduction
Safety in work zones is an important consideration to highway agencies. Work zones frequently have
detours, reduced shoulders, reduced lane widths, limited “pull-off ” areas, unusual maneuvering, tempo-
rary pavement markings, rough pavement, and many other conditions that make navigating the work zone
more difficult than completed roadway sections.
Roadway lighting can be an especially effective tool in work zones. Lighting can provide additional
visual information by reducing veiling glare through the mitigation of other light sources and by provid-
ing additional positional information with respect to other vehicles and objects.
Temporary lighting, when used, should be integrated with the traffic control plan.
Cost
The benefit–cost ratio for temporary roadway lighting is generally very high. Costs should be expected
to be similar to permanent lighting. Planned costs for temporary lighting should be with respect to the
length of time that the construction will take place.
Types
Types of temporary lighting include installing the permanent lighting early in the project, installing
temporary fixtures on permanent poles, or installing the permanent poles in temporary locations. If high
mast is used, the high-mast poles can be placed in temporary locations, fitted with temporary light fix-
tures, and powered by temporary overhead wiring.
Design Values
An attempt should be made in the design of temporary roadway lighting to meet or even exceed the
requirements shown in Table 3-5 of Chapter 3. However, it is often not practicable to meet uniformity and
light level requirements because of various factors of construction. In these cases, it is allowable to relax
requirements as necessary.
Since temporary roadway lighting is a safety enhancement and not an aesthetic enhancement, standard
equipment requirements (such as galvanized poles and cutoff fixtures) may also be relaxed.
Safety
Safe roadside design remains an important issue for temporary lighting. The safety benefit of tempo-
rary roadway lighting does not take precedence over safe roadside design. Hence, temporary lighting
should meet all the protection or breakaway requirements that permanent lighting must meet. Breakaway
39
light poles should not be wired overhead unless they are designed with effective breakaway devices (such
devices are not yet developed as of the publication of this guide but may be possible).
Work Area Lighting
An increasing amount of highway construction is being performed at night to avoid daytime conges-
tion. Lighting is a key factor in performing construction at night. Research project NCHRP 5-31 devel-
oped guidelines for work zone lighting for workers and inspectors as well as roadway lighting for
motorists traveling through the work zone. Interested readers are referred to those reports for more infor-
mation. Temporary lighting for work zones should not be installed without considering the glare effect on
motorists and, if necessary, making accommodations for temporary roadway lighting.
40
Chapter 7
Roundabouts
Introduction
Experience shows that roundabouts require special attention with respect to lighting. Motorist unfa-
miliar with the roundabout will need considerably more navigational information than typical intersec-
tions. Depending on geometric factors and landscape factors in the center island, motorists may be faced
with limited sight distances, weaving traffic, extensive direction and other signs, and the need for quick
decisions. Good lighting will improve driver confidence and performance in navigating through the
roundabout.
Warrant Conditions
The warranting conditions in this section are for the purpose of establishing a basis on which lighting
for freeways may be justified. The warrants provide minimum conditions to be met whenever an agency
is contemplating lighting for new or existing facilities. Meeting of the warrants does not obligate the
highway agency to provide lighting. Jurisdictions may adopt higher numerical warranting values or more
restrictive subjective values for local use.
The decision to provide lighting, and the extent of that lighting once warranting conditions are met,
rests with the highway agency. In determining when and where lighting is to be provided, the agency may
adopt a policy that includes these warrants and is also based on such factors as availability of funds, traf-
fic and crash data, roadway safety rating, etc. The warrants are not to be construed as the only criteria for
justifying lighting. The warrants are intended to be an easily understood tool to assist administrators and
designers in considering lighting on roadways.
Local conditions, such as frequent fog, ice, snow, roadway geometry, ambient lighting, sight distance,
signing, etc., could justify modification of these warrants.
Lighting for roundabouts is considered justified for any roundabout where the local governmental
agency finds sufficient benefit in the forms of convenience, safety, policing, community promotion, pub-
lic relations, etc. to pay an appreciable percentage of the cost of or wholly finance the installations, main-
tenance, and operation of the lighting facilities.
Recommended Lighting Level
Roundabouts should be lit to a level that is 1.3 to 2 times the values used on the best lit approach.
Uniformity should be 3:1 or better. The illuminance method should be used. These light levels should also
be extended 6 to 10 feet outside the outer curb to the sidewalk or other roadside features.
The lighting should be extended a minimum of 400 feet along each road connecting to the roundabout.
Light levels on these should meet the values shown in Table 3-5 or as otherwise required.
Providing good pedestrian recognition is an important issue at roundabouts. Crosswalks at round-
abouts should typically be lit with the pedestrians in positive contrast. Light poles placed 10 to 30 feet
before the crosswalk is recommended for this purpose. Roundabouts should be lit from the outer edge of
the roadway will also aide in providing this positive contrast to pedestrians. Other typical light pole rec-
ommended locations are also shown in Fig 7-1.
41
Figure 7-1. Typical light pole locations.
Many extenuating circumstances, such as brightly lit objects in the center medians (often fountains),
alternative pedestrian crosswalk identification, landscape lighting, retroreflective panels, etc., may alter
these recommendations. Lighting engineers should study the motorist visual field carefully and design
the lighting system to coordinate with all the design elements of the roundabout.
42
Chapter 8
Electrical Systems Requirements
Disconnects
Breakaway rated fuse holders should be used anytime breakaway pole devices are used. Several types
of wiring devices from various manufacturers suitable for this purpose have been in use for some time
with a great deal of success.
Breakaway poles should not be wired from overhead, unless some adequate means of electrical dis-
connecting is designed.
Guidelines Available
The AASHTO publication entitled Standard Specifications for Structural Supports for Highway Signs,
Luminaires, and Traffic Signals has guidelines for breakaway capability and stub height remains.
Requiring electrical disconnects to break away as close to the foundation as possible is not applicable to
all circumstances. While it may be a good idea to limit the distance that a pole may move prior to dis-
connecting the wiring, this “as close as possible” requirement is overly restrictive and should not be con-
sidered mandatory.
Grounding
Equipment should be grounded in accordance with the National Fire Protection Association’s publica-
tion titled, National Electrical Code, including the installation of an equipment grounding conductor.
Ground rods as the sole return path are not adequate. All metal parts of the raceway should be connected
to the grounding conductor including metal ground box lids, exposed metal conduit, metal poles, and sup-
plemental ground rods at pole foundations (and other locations).
Voltage Drop
The voltage available at each light pole affects the light output of the fixture. Voltage drop values
should be considered in the determination of lumen maintenance factors. This may be alleviated with the
use of electronic ballasts, which could offer better light output regulation and lower losses. Electronic bal-
lasting of HID light sources is anticipated to be available in the near future, and it is expected to be worth
the increased costs.
43
Chapter 9
Safety Rest Areas
OVERVIEW
Rest areas typically offer complete rest facilities, i.e., including comfort station and picnic area. By
their very nature, these facilities incorporate both vehicular and pedestrian usage, and constitute an
important highway feature to the traveling public. They are available for use at night as well as by day,
and their general appearance should generate a feeling of safety and security. This condition can exist
only if the facility is adequately lighted for nighttime use. An exception may be made for all or part of rest
areas where darkness is desired, such as for sleeping, for enjoyment of the natural nighttime environment,
or for rural areas.
Properly designed lighting, conventional or high mast, will enhance the architectural and landscape
features of the facility, promote safety by easing the task of policing, and contribute to the rest and relax-
ation of the motorist.
Warrant
Any rest area offering complete rest facilities may be lighted.
Design Values
The overall design of the lighting has been divided into general areas as follows:
• entrance and exit
• interior roadways
• parking areas
• activity areas
• main lanes
(For a summary of recommended lighting criteria, see Tables 9-1 and 9-2.)
These have been defined for separate consideration as each is to be used for a specific and different
purpose. The design, however, should also consider the interrelationship of all of these areas.
The light values recommended in this section are the average maintained values as previously defined
in this guide. The following guidelines should be used except as noted at the beginning of this chapter.
Entrance and Exit
These are defined as the deceleration and acceleration lanes adjacent to the main roadway, leading to
and from the gore areas.
The entrance and exit lanes may be lighted so that the driver desiring to enter or leave the rest area can
safely make the transition from the main roadway to the rest area and vice-versa.
Lighting should be provided along the deceleration lane. An average maintained illuminance of
0.6 footcandle (6 lux), or a luminance of 0.4 candela per square meter (0.12 footlambert), should occur at
45
TABLE 9-1. Recommended Maintained Lighting Levels for Rest Areas—Illuminance
Location Footcandles Lux Uniformity Ratio
Entrance and Exit Gores* and Main Lanes 0.6 6 3:1 to 4:1
Interior Roadways 0.6 6 3:1 to 4:1
Parking Areas 1.0 11 3:1 to 4:1
Active Areas:
• Major 1.0 11 3:1 to 4:1
• Minor 0.5 5 6:1
* The illuminance values for entrance and exit gores and for interior roadways are for R3 surfaces. If an R1 surface is
used, the values may be reduced by approximately 25 to 30 percent.
This table assumes a rural setting. For rest areas in urban settings, use the values of Table 3-5.
the gore point between the deceleration lane and the beginning of the interior roadways. This is based on
the use of from three to five luminaire locations along the length of the speed change lanes.
Similarly at the exit, an average maintained illuminance of 0.6 footcandle (6 lux) or a luminance of
0.4 candela per square meter (0.12 footlambert) should occur at the exit gore.
In the event that the main roadway is continuously lighted beyond the confines of the rest area, decel-
eration and acceleration lanes should be lighted to a level equal to that of the main roadway.
Interior Roadways
These are roadways between the entrance gore point and the parking areas and from the parking areas
to the exit.
When lighting is provided, it is recommended that the average maintained illuminance should be
0.6 footcandle (6 lux) or a maintained luminance of 0.4 candela per square meter (0.12 footlambert). This
continues the lighting level obtained at gore points.
TABLE 9-2. Recommended Maintained Lighting Levels for Rest Areas—Luminance
Veiling Luminance
Lavg Uniformity Ratio*
Foot
Location cd/m2 Lamberts Lavg/Lmin Lmax/Lmin Lv(max)/Lavg
Entrance and Exit Gores 0.4 0.12 3.5:1 6:1 0.3:1

Interior Roadways 0.4 0.12 3.5:1 6:1 0.3:1
Parking Areas &
Use Illuminance Method
Activity Areas
* The above uniformity ratios are the maximum allowable. Lower numerical ratios produce better uniformity and are
desirable.
This table assumes a rural setting. For rest areas in urban settings, use the values of Table 3-5.
46
Safety Rest Areas
Parking Areas
Lighting of the parking areas, both automobile and truck, should be such that the motorist, while still
in his vehicle, can distinguish features of the area, as well as discern pedestrians moving about the area.
An average maintained lighting level of 1.0 footcandle (11 lux) with a uniformity ratio of 3:1 to 4:1
should be used over all parking facilities.
Attention should be applied to special areas, such as handicap ramps, sanitary disposal stations, and
other items, which may require special detailing. This may be done by placing a luminaire in close prox-
imity to a particular spot so that maximum visibility may be obtained by special task lighting or other spe-
cial treatment.
Activity Areas
The activity areas are those designed for pedestrian use.

The major activity areas are those which include such facilities as rest rooms, information centers, etc.,
as well as the walkways to and from these locations and to the parking area.
The minor activity areas are those which include picnic tables, dog walks, etc., and their associated
walkways and facilities.
It is recommended that the main walkways around the structures and major walkways leading to and
from the parking facilities be lighted to 1.0 average maintained footcandle (11 lux) with a 3:1 to 4:1 uni-
formity. This is in keeping with the recommendation for the parking areas.
Those walks leading to the shelter table, picnic tables, dog walks, etc., should be lighted to 0.5 aver-
age maintained footcandle (5 lux) utilizing a 6:1 or better uniformity ratio.
Main Lanes
Lighting the area adjacent to the roadway without also lighting the roadway will create a veiling glare.
Since the rest area is a part of the highway facility and under the jurisdiction of the governing author-
ity, this veiling glare should be mitigated. The main lanes through this area therefore should also be
lighted unless the rest area is spaced a sufficient distance from the main lanes or otherwise blocked from
view so as not to cause an adverse veiling glare for main lane motorists.
47
Chapter 10
Roadway Sign Lighting
10.1 OVERVIEW
Introduction
Traffic signs are placed along the roadway in strategic locations and are used to convey specific, con-
sistent messages to the motorist. The standards used in the design of traffic signs are discussed in the
Manual on Uniform Traffic Control Devices (MUTCD). The intent of these standards is to ensure that all
traffic signs are designed to provide a message that can quickly and accurately convey the necessary
information and to provide consistency in the design of signs nationwide.
Sign legibility at night can be achieved in one of two ways:
• the retro-reflection of the letters and background of the sign by vehicle headlights
• the illumination of the sign face by an internal or external fixed-source sign lighting system.
The retro-reflection concept is used almost universally for traffic signs place along the edge of the
roadway. The illumination of the sign face by a fixed-source sign lighting system has generally been used
on traffic signs placed over the roadway or in cases where the retro-reflective properties of the sign will
not provide adequate legibility.
A sign designed to be legible under daylight conditions can be illuminated to fulfill its basic purpose
at night. A properly designed sign lighting system can aid the motorist with the rapid and accurate recog-
nition of the sign’s shape, color, and message. This serves to improve safety by reducing the possibility
that motorists will stop or significantly reduce their speed at locations where signs may be otherwise dif-
ficult to read. Sign lighting can also be used on signs installed in locations having a high level of ambient
light in order to make the sign legible against the surrounding area.
Key Elements of Roadway Sign Lighting
The following elements of roadway sign lighting are key in making the sign message visible and
legible to the passing motorist during the hours of darkness:
• Ambient Luminance
The amount of ambient luminance adjacent to the sign determines the amount of sign lighting
required to accurately convey the sign message to passing motorist. If the amount of ambient lumi-
nance is high, then sign lighting can be provided to make the sign legible against the surrounding
area.
• Sign Luminance Above Ambient
The sign luminance in excess of the ambient luminance determines how well the sign can be viewed
against the bright background. When the difference between sign luminance and ambient luminance
increases, the ability of a passing motorist to rapidly and accurately recognize the shape and color of
the sign as well as the sign message against the competing distractions in the surrounding area also
increases.
49
• Uniformity Ratio of Light Levels

A suitable level of uniformity of light over the entire face of the sign will add to its effectiveness and
ability to accurately convey the sign message to the passing motorist. In performing sign lighting
calculations, the maximum-to-minimum uniformity ratio has been established as the standard means
by which the uniformity of light levels on the face of a sign is to be determined.
• Reflectivity of Legend and Background
The reflectivity of the letters that make up the legend as well as the background that they are
installed on should be optimized without delivering excessive glare to the passing motorist. The
color of the sign and the retro-reflective material utilized will control the amount of light that is
reflected. The standards that dictate the colors of a sign are discussed in the MUTCD.
• Contrast Between Legend and Background
The contrast between the letters that make up the legend and the background that they are installed
on should be optimized in order for a passing motorist to quickly read and process the sign message.
The color of the sign and the retro-reflective material utilized will control the amount of contrast that
is achieved. The standards that dictate the colors of a sign are discussed in the MUTCD.
The above elements will be described in more detail in the sections that follow. An evaluation of each of the
above elements should be made during the design of the sign lighting system in order to ensure that the
sign message is adequately conveyed to the motorist during the hours of darkness.
10.2 SIGN LIGHTING WARRANTS
The standards that are used as a basis for warranting the installation of sign lighting are discussed in
the MUTCD. The MUTCD states “signs shall be retro-reflective or illuminated to show the same shape
and similar color by both day and night.” Therefore, sign lighting is warranted when the retro-reflective
properties of the sign face alone will not perform effectively or provide sufficient legibility without exter-
nal sources of nighttime illumination.
Although there are many materials currently available for the retro-reflection of signs as well as for the
illumination of signs, it is important to keep in mind that the standard requirements for daylight sign color
should be maintained by nighttime illumination. Sign lighting can also be used on signs that are made
with retro-reflective materials in order to enhance their visibility. This becomes especially important in
areas that exhibit frequent and severe conditions of condensation or frost or in areas where there is a high
level of dirt particles in the atmosphere that could accumulate on the sign face.
In an effort to reduce utility costs for electrical energy and subsequent maintenance costs associated
with sign lighting systems, the retro-reflection properties of signs alone may perform effectively without
the need for additional nighttime illumination. However, the following conditions should be satisfied:
1. The sign is in an area that contains a low-to-intermediate ambient light level, and
2. There is at least 1200 feet (366 meters) or more of tangent sight distance in advance of the overhead
sign.
Exceptions to the above conditions that may warrant the installation of sign lighting to overhead or
ground mounted signs are:
1. The signs are diagrammatic.
2. The signs are for “Exit Only” lane drops.
3. The signs are contained within a high-volume interchange.
50
10.3 ILLUMINATED SIGN TYPES
Signs can be illuminated in a variety of different ways in order to make the sign message visible and
legible to the passing motorist during the hours of darkness. The three main ways of providing illumina-
tion to a sign are as follows:
1. Externally Illuminated
Externally illuminated signs are traffic signs that are illuminated by a source of light that is
mounted external to the sign and the sign message becomes understandable when uniformly illu-
minated by this source.
2. Internally Illuminated
Internally illuminated signs are traffic signs that are illuminated by a source of light that is enclosed
within the sign and the sign message becomes understandable when illuminated from within
because of the difference of color and transparent nature of the material that makes up the sign face.
3. Luminous Source Message
Luminous source message signs are dynamic signs that formulate messages by lighted pixels or
characters against a dark or black background. They electronically vary the sign display and provide
the motorist with real-time information on current traffic, roadway, or environmental conditions.
This type of sign also provides the greatest amount of flexibility and control because the sign mes-
sage can be changed virtually instantaneously from a remote location as traffic conditions warrant.
Luminous source message signs are becoming more widely used due to their ability to compensate for
a multitude of environmental and roadway conditions. Various types of luminous source technologies are
available in both non-matrix and matrix formats.
• Non-matrix format signs are often called “blank-out” signs and offer a fixed selection of messages.
The sign messages are displayed by activating the appropriate light source and extinguishing all others.
• Matrix format signs use matrices of pixels where each pixel represents an individual light source.
The sign messages are displayed by activating the appropriate pixels that are used to form the indi-
vidual characters and extinguishing all others.
Since roadway lighting primarily focuses on fixed-source lighting, the discussion that follows in this
guide will cover externally illuminated signs only.
Ambient Luminance Classifications
The amount of ambient light in the area adjacent to where a roadway sign is present plays an impor-
tant role in determining the amount of sign lighting that is required in order to adequately convey the sign
message to the motorist at night. This ambient light is called the ambient luminance and is defined as the
background luminance against which a sign is to be viewed by a passing motorist. High levels of ambi-
ent luminance may make sign lighting warranted regardless of the retro-reflective properties of the sign
face material. Since there is no approved method for measuring ambient luminance, the three classifica-
tions of ambient luminance that are listed below have been developed as a guide for the lighting engineer:
1. Low
Low levels of ambient luminance exist in rural areas without roadway and/or intersection lighting.
Objects at night are visible only in bright moonlight. There is very little or no other lighting in the area.
2. Medium
Medium levels of ambient luminance exist in intermediate areas with some roadway and/or inter-
change lighting. May contain small areas of commercial lighting.
51
3. High
High levels of ambient luminance exist in urban areas with high levels of roadway lighting. May
contain brightly lighted commercial advertising signs, building facades, and/or highly illuminated
parking facilities.
10.4 SIGN LIGHTING RECOMMENDATIONS
Once it has been determined that sign lighting is warranted, the lighting engineer should select a light
source that will light the sign so that it exhibits the same color rendering properties during the hours of
darkness as it did under daylight conditions. The amount of light that is required to adequately light the
sign during the hours of darkness is defined as the quantity of light.
There are several different types of light sources available from which to light roadway signs. Each
light source has its own set of unique characteristics that may make it more desirable than others for a
given sign installation. Energy consumption is a major consideration in choosing a light source and
should be considered. However, there are other factors such as color rendering, operating temperature,
efficiency, and ease of maintenance that are equally important and should also be evaluated.
The light source that is selected should be able to adequately light the face of the sign without inter-
fering with the contrast between the letters that make up the legend and the background of the sign that
they are installed on. The contrast between the letters and the background will determine how quickly and
accurately a passing motorist can recognize the shape and color of the sign as well as the interpretation of
the message that is being displayed.
The amount of sign lighting that is required in order to adequately convey the sign message to the
motorist at night is also dependent on the amount of ambient luminance in the area adjacent to where the
sign is present. The recommended average maintained levels of illuminance or luminance for the three
classifications of ambient luminance are shown in the table below.
TABLE 10-1 Illuminance and Luminance Levels for Sign Lighting*
Ambient
Luminance Sign Illuminance Sign Luminance**
Candelas per Candelas per

Footcandles Lux Square Meter Square Foot
Low 10–20 100–200 22–44 2.2–4.4

Medium 20–40 200–400 44–89 4.4–8.9
High 40–80 400–800 89–78 8.9–17.8
* Adapted from The IESNA Lighting Handbook, Reference & Application, 9th Edition, Illuminating Engineering Society
of North America. Used by permission.
** Based on a maintained reflectance of 70 percent for white sign letters.
The lighting engineer may utilize either the illuminance or luminance recommendations that are pro-
vided in the above table.
52
Lighting Uniformity
Uniformity of lighting is an indication of the quality of illumination and can be defined as either the
average-to-minimum, maximum-to-minimum, or maximum-to-average ratios of light levels that are pre-
sent on the face of the sign. In performing sign-lighting calculations, the maximum-to-minimum ratio has
been established as the standard means of determining the uniformity of light levels that appear on the
face of a sign.
The uniformity of the light levels that appear on the face of the sign should be controlled if the sign is
to be effective in conveying the sign message to the motorist at night. Suitable uniformity over the entire
face of the sign will provide consistent and proportional contrast that is similar to daytime conditions.
Maximum and minimum points that are spaced too close together will provide poor contrast between the
letters that make up the legend and the background of the sign making it more difficult to read.
A maximum-to-minimum uniformity ratio of 6 to 1 is recommended as an acceptable ratio of lighting
levels on the face of the sign. Since lower ratios will produce a more pleasing appearance and a more leg-
ible sign, lower maximum-to-minimum uniformity ratios are preferred. Because the mode of reflectance
of the sign material is assumed to be mostly diffuse, this recommended uniformity ratio would apply
whether illuminance or luminance values are used.
Sign Color Standards
The standards that are used as a basis for sign colors are discussed in the MUTCD. The sign colors
have been standardized nationwide so it is essential that the face of the sign be properly illuminated in
order to retain the colors for identification purposes. The lighting engineer should verify that the color
rendering properties of the illuminated sign are not degraded from that viewed during daylight conditions.
Placement of Lighting Units
The lighting units that illuminate the face of a sign may be located on either the top of the sign, the bot-
tom of the sign, or may be remotely located on an adjacent support. The lighting engineer should evalu-
ate the following considerations before selecting the mounting arrangement that is to be utilized.
1. The luminaire housing should not obstruct the view of the sign message.
2. The reflected light should not reduce the visual performance of the sign message
3. Contribution to sky-glow should be limited as much as is practicable.
4. The spill light should not be directed into the eyes of motorists.
5. The luminaire mounting arrangement should not create maintenance problems.
Locating the lighting units on the bottom of the sign, if practical, is generally the preferred alternative
for the following reasons.
1. The reflected light is less likely to reduce the visual performance of the sign message or produce
reflected glare into the eyes of motorists.
2. The lighting units do not produce daytime shadows and reflections from the sun on the face of the sign.
3. The lighting units are easier to access for maintenance.
4. The lighting units may collect snow and dirt, but may also be cleaned by rain.
5. The face of the sign may only partially shield the light that spills onto traffic approaching from the
rear of the sign. However, a separate shielding mechanism can be provided on the lighting units that
will minimize this effect.
6. Excess sky-glow or light pollution may be inherent. However, a separate shielding mechanism can
be provided on the lighting units or optical control equipment can be utilized in order to minimize
these effects.
53
7. The lighting units may obstruct the view of the sign message at some viewing angles. However,
proper placement and installation of the lighting units can minimize this problem.
In addition to the above considerations, the lighting engineer should also verify that the adjacent road-
way lighting system, if present, does not adversely impact the lighting levels on the face of the sign or
physically block the face of the sign. The adjacent roadway lighting system is not intended to perform the
lighting of the adjacent overhead retro-reflective signs.
10.5 REFERENCES
1. Illuminating Engineering Society of North America, IESNA RP-19-01, Roadway Sign Lighting.
IESNA, New York, NY, March 2001.
2. Illuminating Engineering Society of North America, The IESNA Lighting Handbook, Reference and
Application, 9th Edition. IESNA, New York, NY.
3. Federal Highway Administration, Manual on Uniform Traffic Control Devices (MUTCD). FHWA,
Washington, DC, 2003.
4. Federal Highway Administration, Traffic Control Systems Handbook. USDOT, FHWA, Washington,
DC, February 1996, Report No. FHWA-SA-95-032.
54
Chapter 11
Maintenance Considerations
in Roadway Lighting Design
General
All lighting systems depreciate with time and need continuing surveillance and maintenance to pro-
vide the service for which they were designed and installed. Lamps, luminaires, support structures, and
other electrical system components cannot be expected to stay efficient and effective without a well-
planned maintenance program. The lighting systems in public facilities, whether they are small with a few
light fixtures or they consist of many high-mast poles in a freeway interchange, are a significant invest-
ment of limited public resources. Good maintenance is warranted as a protection of that investment. It
should be kept in mind that the lighting systems on the roadway, while not required for safety, may
enhance nighttime safety.
The design of lighting should consider the extent and frequency of maintenance, which will be pro-
vided by operating agency. In the initial design stage, the designers should select and specify quality
products with high reliability and long life as much as practicable to achieve favorable conditions for the
system operation and maintenance workers.
Lighting maintenance can be categorized into several basic areas including luminaires, support struc-
tures, electrical distribution and control, and external factors. Each of these areas is important to the over-
all utility and efficiency of a lighting installation and should be included in the planning of any good
maintenance program.
Maintenance Factors
Luminaire Dirt Depreciation (LDD)
The general accumulation of dirt due to airborne particles deposited on the surface of the lamp and
luminaire and deterioration of optical material (luminaire lens), reduce luminaire lumen output and
change the distribution characteristics on the roadway. It is necessary for the lighting designer to estimate
the effect of dirt accumulation and to allow for this depreciation in the design values. The selection of
luminaire type, mounting height, environment of the luminaire location (urban or rural setting), traffic
volume and roadway offsets will affect the rate of dirt depreciation. Technical information included in the
references in this design guide and in other published data will provide more information on minimizing
dirt accumulation on luminaires. This information can be used to determine the frequency of luminaire
cleaning to maintain the lighting system close to the designed lighting values.
Lamp Lumen Depreciation (LLD)
The gradual loss of lumen output of light source due to normal in-service aging, varies depending on
the type of light source. Manufacturers’ published data for each lamp type and size offers statistical
information which can be used to predict the rate of loss in lamp lumen output as well as to estimate
lamp failure rates. These predicted output losses and life expectancies should be used in developing a
maintenance plan. It is important to use this maintenance schedule to keep the lighting system above the
55
minimum values established in the lighting design and to achieve the most favorable economy of lamp
operation. Group replacement of some types of lamps near the end of their expected life may result in
the lowest overall cost of lamp replacement and provide the most effective service to the public.
Equipment Factor (EF)
Equipment factor is a general factor encompassing luminaire losses due to all other factors such as
ballast factor, manufacturing tolerances, voltage drop, lamp position, ambient temperature, and lumniaire
component depreciation.
For design values in lighting levels as presented in this design guide, designers should consider
appropriate reduction in the initially calculated values. The reduction factor, referred to as the total LLF
(Light Loss Factor, See the online appendices at http://downloads.transportation.org/lighting.pdf), is a
combination of LDD (Luminaire Dirt Depreciation), LLD (Lamp Lumen Depreciation), and EF
(Equipment Factor, including voltage drop). Values in the range of 60 to 80 percent (of initial design
value) are used for high-pressure sodium (45 to 65 percent for Metal Halide) general application such
as regularly maintained outdoor luminaires installed on lighting poles. The use of realistic luminaire
depreciation, dirt, and equipment factors, is essential in lighting design to achieve the expected lighting
levels on the roadway after the lighting system is installed. Values for these factors are obtained from
manufacturers’ product data, and engineering judgement based on the designer’s experience.
Support Structure Maintenance
The selection of luminaire mounting heights, location, and material of supports and their foundations,
should be made with a full knowledge of maintenance practices and costs as well as structural design
practice by the responsible agency. An important consideration in facilitating maintenance activities is to
provide for the safety of maintenance personnel and ease of access to the lighting system. Lighting equip-
ment should be located where it can be serviced with the minimum exposure of the work crew and mini-
mum inconvenience to the motoring public.
Prior to the selection of poles, brackets, junction boxes and other equipment which will be exposed to
the weather, it is essential to consider both the initial cost and the long-term cost of maintenance. The day-
time appearance of a lighting installation may be important in the view of the public. A poorly maintained
and unsightly appearing lighting installation can become a source of complaint of the public to the
responsible agency, and can leave unfavorable remarks on the agency’s overall public service.
Electrical Distribution and Control System Maintenance
A variety of distribution and control systems are used in roadway lighting. It is not the purpose of this
design guide to prescribe specific types of cable or control equipment for roadway lighting. The designer
should consider the necessary control functions, and make careful evaluation on available control sys-
tems for their advantages and long-term maintenance requirements. In terms of maintenance costs and
dependability of operation, generally the simplest form of control equipment that performs the desired
function is preferred. Complex control and distribution systems increase the potential for failure and
require more frequent attention by maintenance personnel. In areas where vandalism is anticipated,
equipment should be designed to the extent practicable to preclude intentional damage attempts and pre-
vent persons from coming in contact with energized electrical components.
56
Maintenance Considerations in Roadway Lighting Design
External Factors
Many external factors can have a significant effect on the maintenance of lighting. Some of these are
listed as follows:
• Low hanging trees located close to luminaires can disrupt light distribution on the roadway, render-
ing the system ineffective, and possibly confusing motorist with shadows on the roadway. The prun-
ing of trees may be a necessary as a part of lighting maintenance activities.
• High dirt conditions, as in industrial areas and roadways with high truck traffic volumes, require
more frequent cleaning of luminaires.
• Corrosive atmospheres may dictate the selection of luminaires, poles, brackets, and other material in
the lighting system that can resist these influences.
• Excessive vibration of luminaires and their supports may occasionally be experienced, particularly
on bridge structures where a combination of wind and mechanical structure movement due to
vehicle traffic can damage the lighting units. The use of vibration dampers or auxiliary supports
and prudent selection of light pole locations on the bridge structure, have been effective in reduc-
ing these forces to tolerable limits.
57
Chapter 12
Sky Glow and Light Trespass
12.1 OVERVIEW
Introduction
Departments of transportation should be aware of how lighting systems affect the area surrounding the
roadway. Across the country, momentum is growing to recognize the objectionable side effects of road-
way lighting as a problem and to address the situation directly. Unless lighting engineers are sensitive to
these issues and achieve some measures of abatement, regulations may be implemented that do not
always provide for optimum roadway lighting.
In an effort to resolve the objectionable side effects of outdoor lighting in general, many communities
are not adopting outdoor lighting ordinances. Some of these rules specify measurable limits for allowable
trespass of lighting, in terms of horizontal or vertical illuminance at or within property lines. Others dic-
tate pole heights or luminaire lamp power (wattage). While municipal lighting ordinances do not usually
apply to work by departments of transportation, such ordinances do indicate a public desire for quality
lighting. Furthermore, more and more states are considering state-level limitations on outdoor lighting, or
may authorize limitations within municipalities.
Outdoor lighting ordinances can actually be very beneficial to improving visibility on roadways, by
limiting the impacts of off-roadway lighting (principally from commercial sites or municipal facilities
such as parks) on drivers. Implementing Master Lighting Plans, coordinated with local citizen input, may
also be used to address objectionable effects of outdoor lighting systems. See the section on Master
Lighting Plans.
Types of Objectionable Roadway Lighting
Four terms often used to described effects from roadway lighting that may be considered objectional
be are shown below:
1. Sky Glow
Sky glow is defined as the added sky brightness caused by the scattering of light into the atmosphere.
Unless absorbed, light emitted by an outdoor luminaire will propagate through the atmosphere.
Depending on atmospheric conditions, some or even most of that light will be scattered. That portion of
scattered light that is redirected back towards the ground is, in essence, light that is emitted by the sky. At
high enough levels, the sky will appear as a self-luminous body, and will glow. Sky glow is of great con-
cern to astronomers and others who like to see the moon and stars, or just wish to enjoy the natural night-
time environment. It is an unavoidable side effect of outdoor lighting, however, much can be done to
minimize the impact.
Recent research in Europe and the United States indicate that most of the light contributing to sky glow
does not come from luminaires with a small percent of output lumens at or above the 90 degree plane, as
was originally thought. A much larger amount of flux is reflected from the roadway and other surfaces
than the amount directly emitted above the horizontal plane by the luminaire. This indicates that the use
of “full-cutoff” luminaires is not automatically better than the use of a cutoff or even semi-cutoff lumi-
59
naires with a beam pattern that best utilizes luminous flux to meet drivers’ needs. In all cases, alternative
lighting designs should be evaluated in order to achieve the best balance between the needs of the public,
while minimizing sky glow and energy consumption.
2. Light Trespass
The term light trespass describes light that strays from its intended target and illuminates adjacent prop-
erties. Most complaints about this impact are from the public, when lighting from roadway luminaires
shines into their windows or onto their property.
3. Glare
In the context of this discussion, glare is defined as when some portion of the emitting or reflecting sur-
face of the luminaire is directly observable by drivers, and the resulting illuminance at the eye of the
driver results in either discomfort or disability glare. Poor lighting designs that result in glare may sig-
nificantly reduce, or possibly eliminate, the positive benefits of the lighting system. To a lesser extent,
the public may complain of glare on private property due to light trespass from roadway lighting lumi-
naires. In most cases, this impact does not actually result in discomfort or disability glare, but is an
annoyance.
4. Visual Clutter
Visual clutter is a term used to describe a subjective evaluation of a visual field in which there are numer-
ous light sources. The appearance of many luminaires in the field of view, while insufficient to have sig-
nificant impacts on visibility, may be distracting or even confusing to drivers, and may reduce a persons
ability to appreciate the nighttime environment.
12.2 MITIGATING SKY GLOW AND LIGHT TRESPASS
The least expensive and most successful approach to objectionable light problems is prevention. For
prevention efforts to work the designer should:
• Perform a review or walk-through of the site during the pre-design stage. This should include an
analysis of adjacent property and nearby developments as well as an investigation into the commu-
nity desires for lighting systems.
• Select a luminaire whose candela distribution pattern matches the need. Establish the position of the
luminaire precisely. If calculations then show the light distribution will be objectionable, make mod-
ifications to wattage or mounting height, or select another luminaire.
• Consider internal and external shields if necessary to limit the candela in certain directions. Also
consider that internal and external shields will alter the photometric distribution of the fixture and
account for such alterations in the lighting system design.
• Consider pole location, mounting height, spacing, finished terrain, and landscaping as design vari-
ables that can be used to mitigate light trespass. Take advantage of natural and man-made obstacles
such as tree lines and retaining walls.
• Remember that glare or visual clutter can be produced by almost any luminaire when observed
against a dark background. Choose luminaires and placements with care.
60
Glossary
absorptance
The ratio of the flux absorbed by a medium to the incident flux.
NOTE: The sum of the hemispherical reflectance, the hemispherical transmittance, and the absorp-
tance is one.
adaptation
The process by which the retina becomes accustomed to more or less light than it was exposed to
during an immediately preceding period, resulting in a change in the sensitivity of the eye to light.
atmospheric transitivity
The ratio of the directly transmitted flux incident on a surface after passing through unit thickness of
the atmosphere to the flux that would be incident on the same surface if the flux had passed through
a vacuum.
average initial illuminance
The average level of horizontal illuminance on the pavement area of a traveled way at the time the
lighting system is installed when lamps are new and luminaires are clean; expressed in average foot-
candles (lux) for the pavement area.
average maintained illuminance
The average level of horizontal illuminance on the roadway pavement when the output of the lamp
and luminaire is diminished by the maintenance factors; expressed in average footcandles (lux) for
the pavement area.
ballast
A device used with an electric-discharge lamp to obtain the necessary circuit conditions (voltage,
current, and waveform) for starting and operating.
bike lane
Any facility that explicitly provides for bicycle travel.
bracket (mast arm)
An attachment to a lamp post or pole from which a luminaire is suspended.
candela (cd)
(formerly candle)
The unit of luminous intensity.
candela per square meter
The International System (SI) unit of luminance (photometric brightness) equal to the uniform lumi-
nance of a perfectly diffusing surface emitting or reflecting light at the rate of one lumen per square
meter, or the average luminance of any surface emitting or reflecting light at that rate. One candela
per square meter equals 0.2919 footlambert.
candlepower (cp)
Luminous intensity expressed in candelas (not an indication of total light output).
61
candlepower distribution curve

A curve, generally polar, representing the variation of luminous intensity of a lamp or luminaire in a
plane through the light center.
central (foveal) vision
The seeing of objects in the central or foveal part of the visual field, approximately two degrees in
diameter. It permits seeing much finer detail than does peripheral vision.
CIE
International Committee on Illumination. Publishers of lighting references and design guides.
coefficient of utilization (CU)
The ratio of the luminous flux (lumens) from a luminaire received on the surface of the roadway to
the lumens emitted by the luminaire’s lamps alone.
complete interchange lighting
The lighting of the freeway through traffic lanes through the interchange, the traffic lanes of all
ramps, the acceleration and deceleration lanes, all ramp terminals, and the crossroad between the
outermost ramp terminals.
constant-current transformer
A device (sometimes erroneously referred to as a constant-current regulator) that automatically
maintains a constant current in its secondary circuit under varying conditions of load impedance
when supplied from a constant potential source.
contrast sensitivity
The ability to detect the presence of luminance differences. Quantitatively, it is equal to the recipro-
cal of the contrast threshold.
contrast threshold
The minimal perceptible contrast for a given state of adaptation of the eye. It also is defined as the
luminance contrast detectable during some specific fraction of the times it is presented to an
observer.
cutoff
Designation for luminaire light distribution limiting up lights. See IESNA classifications.
diffuse reflectance
The ratio of the flux leaving a surface or medium by diffuse reflection to the incident flux.
diffuser
A device to redirect or scatter the light from a source, primarily by the process of diffuse transmission.
disability glare
Glare resulting in reduced visual performance and visibility—often accompanied by discomfort. See
veiling luminance.
discomfort glare
Glare producing discomfort. It does not necessarily interfere with visual performance or visibility.
Equipment Factor
A factor used in illuminance or luminance calculations to compensate for light losses due to normal
production tolerances of commercially available luminaires when compared with laboratory photo-
metric test models.
fixture
See luminaire.
62
Glossary
footcandle (fc)
The unit of illumination when the foot is taken as the unit of length. It is the illumination on a sur-
face one square foot in area on which there is a uniformly distributed flux of one lumen, or the illu-
mination produced on a surface, all points of which are at a distance of one foot from a directionally
uniform point source of one candela.
footlambert (fL)
A unit of luminance (photometric brightness) equal to 1/π candela per square foot, or to the uniform
luminance of a perfectly diffusing surface emitting or reflecting light at the rate of one lumen per
square foot, or to the average luminance of any surface emitting or reflecting light at that rate. See
luminance and candela per square meter.
glare
The sensation produced by luminance within the visual field that is sufficiently greater than the
luminance to which the eyes are adapted to cause annoyance, discomfort, or loss in visual perfor-
mance and visibility. See disability glare and discomfort glare.
NOTE: The magnitude of the sensation of glare depends on such factors as the size, position, and
luminance of a source, the number of sources, and the luminance to which the eyes are adapted.
high-intensity discharge lamps
A general group of lamps that includes mercury, metal halide, and high pressure sodium lamps.
high mast lighting
Illumination of a large area by means of a group of luminaires designed to be mounted in fixed ori-
entation at the top of a high mast, generally 80 feet (25 meters) or higher.
high pressure sodium lamp
A sodium vapor lamp in which the partial pressure of the vapor during operation is on the order of
104 N • m22 (0.1 atmosphere).
IESNA
Illuminating Engineering Society of North America. Publishers of lighting references and design
guides.
illuminance
The density of the luminous flux incident on a surface; it is the quotient derived by dividing the lumi-
nous flux by the area of the surface, when the latter is uniformly illuminated.
isocandela line
A line plotted on any appropriate coordinates to show directions in space, about a source of light,
in which the candlepower is the same. For a complete exploration, the line always is a closed
curve. A series of such curves, usually for equal increments of candlepower, is called an isocandela
diagram.
isolux (isofootcandle) line
A line plotted on any appropriate coordinates to show all the points on a surface where the illumina-
tion is the same. For a complete exploration, the line is a closed curve. A series of such lines for var-
ious illumination values is called an isolux (isofootcandle) diagram.
lambert (L)
A unit of luminance (photometric brightness) equal to 1/π candela per square centimeter.
lambert surface
A surface that emits or reflects light in accordance with Lambert’s cosine law. A lambert surface has
the same luminance regardless of viewing angle.
63
lamp lumen depreciation factor (LLD)

The multiplier to be used in illumination calculations to relate the initial rated output of light sources
to the anticipated minimum rated output based on the relamping program to be used. (See “Light
Loss Factor” discussion earlier in Chapter 11).
light center (of a lamp)
The center of the smallest sphere that would completely contain the light-emitting element of the
lamp.
light center length (of a lamp)
The distance from the light center to a specified reference point on a lamp.
lighting unit
The assembly of pole or standard with bracket and luminaire.
Light Loss Factor
A depreciation factor applied to the calculated initial average luminance or illuminance.
low pressure sodium lamp
A sodium vapor lamp in which the partial pressure of the vapor during operation is on the order of
30 millimeters of mercury (0.04 atmosphere).
lumen (lm)
The unit of luminous flux. It is equal to the flux through a unit solid angle (steradiam), from a uni-
form point source of one candela (candle), or to the flux on a unit surface all points of which are at
unit distance from a uniform point source of one candela.
luminaire
A complete fixture consisting of a lamp or lamps together with the parts designed to distribute the
light, position and protect the lamps, and connect the lamps to the power supply.
luminaire efficiency
The ratio of luminous flux (lumens) emitted by a luminaire to that emitted by the lamp or lamps used
therein.
luminance (photometric brightness)
1. Luminance in a direction, at a point on the surface of a source of a receiver, or of any other real
or virtual surface is the quotient of the luminous flux leaving, passing through, or arriving at an
element of the surface surrounding the point, and propagated in directions defined by an elemen-
tary cone containing the given direction, by the product of the solid angle of the cone and the area
of the orthogonal projection of the element on the surface on a plane perpendicular to the given
direction. See candela per square meter.
2. The luminous intensity of any surface in a given direction per unit of projected area of the surface
as viewed for that direction.
NOTE: In the defining equation, θ is the angle between the direction of observation and the direction
normal to the surface.
In common usage, the term brightness usually refers to the intensity of sensation resulting from
viewing surfaces or spaces from which light comes to the eye. This sensation is determined in part
by the definitely measurable luminance defined above and in part by conditions of observation, such
as the state of adaptation of the eye. In much of the literature the term brightness, used alone, refers
to both luminance and sensation. The context usually indicates which meaning is intended.
luminance ratio
The ratio between the luminances (photometric brightness) of any two areas in the visual field.
64
Glossary
luminous efficacy of a source of light

The quotient of the total luminous flux emitted by the total lamp power input. It is expressed in
lumens per watt.
lux (lx)
The International System (SI) unit of illumination. It is the illumination on a surface one square
meter in area on which there is a uniformly distributed flux of one lumen, or the illumination pro-
duced at a surface all points of which are at a distance of one meter from a uniform point source of
one candela.
maintenance factor (MF)
A factor formerly used to denote the ratio of the illumination of a given area after a period of time to
the initial illumination on the same area.
mean lamp lumens
Mean lumen output of a lamp, calculated by determining the area beneath the lumen maintenance
characteristic curve of that source over a given period of time and dividing that area by the time
period in hours.
mercury lamp
An electric discharge lamp in which the major portion of the radiation is produced by the excitation
of mercury atoms.
metal halide lamp
A discharge lamp in which the light is produced by the radiation from a mixture of a metallic vapor
(for example, mercury) and the products of the disassociation of halides (for example, halides of
scandium or sodium).
mounting height (MH)
The vertical distance between the roadway surface and the center of the apparent light source of the
luminaire.
non-cutoff
The luminaire light distribution category when there is no candlepower limitation in the zone above
maximum candlepower. See IESNA classifications.
overhang
The distance between a vertical line passing through the luminaire and the curb or edge of the
roadway.
Lighting consisting of a few luminaires located in the vicinity of some or all ramp terminals, inter-
sections, or other decision-making areas.
pedestrian ways
Public sidewalks for pedestrian traffic generally not within rights-of-way for vehicular traffic road-
ways. Included are skywalks (pedestrian overpasses), subwalks (pedestrian tunnels), walkways giv-
ing access to park or block interiors and crossings near centers of long blocks.
point of fixation
A point or object in the visual field at which the eyes look and upon which they are focused.
rapid start fluorescent lamp
A fluorescent lamp designed for operation with a ballast that provides a low-voltage winding for
preheating the electrodes and initiating the arc without a starting switch or the application of high
voltage.
65
reaction time
The interval between the beginning of a stimulus and the beginning of the response of the observer.
reflector
A device used to redirect the luminous flux from a source by the process of reflection.
refractor
A device used to redirect the luminous flux from a source, primarily by the process of refraction.
semi-cutoff
Designation for luminaire light distribution when the candlepower per 1000 lamp lumens does not
numerically exceed 50 (5 percent) at an angle of 90 degrees above nadir (horizontal), and 200 (20 per-
cent) at a vertical angle of 80 degrees above nadir. This applies to any lateral angle around the luminaire.
spacing
For roadway lighting the distance between successive lighting units, measured along the center line
of the street.
spacing-to-mounting height ratio (S/MH)
The ratio of the distance between luminaire centers to the mounting height above the roadway.
street lighting luminaire
A complete lighting device consisting of a light source together with its direct appurtenances such as
globe, reflector, refractor, housing, and such support as is integral with the housing. The pole, post,
or bracket is not considered part of the luminaire.
street lighting unit
The assembly of a pole or lamppost with a bracket and a luminaire.
transverse roadway line (TRL)
Any line across the roadway that is perpendicular to the curb line.
uniformity of illuminance
The ratio of average footcandles (lux) of illuminance on the pavement area to the footcandles (lux)
at the point of minimum illuminance on the pavement, commonly called the uniformity ratio.
uniformity of luminance
The ratio average level-to-maximum point of luminance or the maximum-to-minimum point. The
average to minimum method uses the average luminance of the roadway design area between two
adjacent luminaires, divided by the lowest value at any point in the area. Maximum-to-minimum
point method uses the maximum and minimum values between the same adjacent luminaires. The
luminance uniformity (avg./min. and max./min.) considers traveled portion of the roadway, except
for divided highways having different designs on each side.
utilization efficiency
A plot of the quantity of light falling on a horizontal plane both in front of and behind the luminaire.
It shows only the percent of bare lamp lumens that fall on the horizontal surface, and is plotted as a
ratio of width of area to mounting height of the luminaire.
veiling luminance
A luminance superimposed on the retinal image that reduces the image contrast. The veiling effect
may be produced by bright sources in the visual field.
visibility
The quality or state of being perceivable by the eye. Visibility may be defined in terms of the dis-
tance at which an object can be just perceived by the eye or it may be defined in terms of the con-
trast or size of a standard test object, observed under standardized view-conditions, having the same
threshold as the given object.
66
Glossary
visual acuity
A measure of the ability to distinguish fine details. Quantitatively, it is the reciprocal of the angular
size in minutes of the critical detail that is just large enough to be seen.
visual angle
The angle subtended by an object or detail at the point of observation. It is usually is measured in
minutes of arc.
walkway
A sidewalk or pedestrian way.
67
Introduction
The appendices contained herein are on-line reerences related to the AASHTO Roadway Lighting
Design Guide, October 2005. Appendix A, Literature Review and References, contains information on
research studies related to accidents and highway lighting, driver parameters, and a summary of further
references. Appendix B, Lighting Basics, provides background information on various issues related to
effective highway lighting.
The AASHTO Roadway Lighting Design Guide can be ordered from the AASHTO Online Bootstore
at http://bookstore.transportation.org or by calling 1-800-231-4375. Please reference code GL-6.
Appendix A
Literature Review and References
(Reprinted from: Evaluation of Roadway Lighting Systems Designed by Small Target Visibility (STV)
Methods, Final Report No. 0-1704-8, Department of Civil Engineering, Texas Tech University,
Lubbock, TX 79409, by Bobby Green, Olkan Culvaci, Sanjaya Senadheera, Karl Burkett, and Douglas
Gransberg)
A.1 CORRELATION BETWEEN ACCIDENTS AND ROADWAY

LIGHTING IMPROVEMENT
Components of Stopping
To understand the correlation between lighting and accidents (crashes), one must first identify
those parameters that affect a driver’s ability to avoid accidents. This is normally expressed through
the components of stopping. To bring a vehicle to a safe stop from some speed, four things must occur
in order:
1. The driver must sample the driving environment for information requiring adjustment in driving
behavior such as changes in speed and direction. This can be called sampling rate and has a proba-
bilistic function associated with it. If a piece of data is sampled which would require a change to
zero, the next three items will occur. This can be called sampling time.
2. The driver must see and acquire an image (for purposes of this discussion, the image will be called
the target) that generates the thought that the vehicle should be stopped. This can be called target
acquisition time.
3. The driver must process that target thought and react by stepping on the brake. This will be called
reaction time.
4. The vehicle must rapidly decelerate from its initial speed to zero. This will be called stopping time.
Stopping time is merely a function of physics and can be computed with great accuracy if the initial
speed is known or can be estimated. Reaction time varies among individuals, but highway safety litera-
ture generally accepts this to be constant at 2.50 seconds. Acquisition time is a more complex parameter
and is a function of both visibility (i.e., the driver being able to see the target) and other more random fac-
tors such as the driver’s immediate attention when the target becomes visible or the driver’s ability to rec-
ognize the target as a hazardous image requiring an immediate reaction. If one were to assume that as the
visibility of the target increases that the probability that an average driver will properly react to it also
increases, then the aim of roadway lighting design for safety should be to create an environment of
enhanced visibility.
Elvik Study
The Norwegian Institute of Transportation Economics conducted a study to validate the hypothesis
that adding light enhances traffic safety (Elvik, 1992). The study looked at the correlation between acci-
dents and roadway lighting in 37 different studies in 11 different countries. The study identified three
types of traffic environment as urban, rural, and freeways and grouped safety data according to these
A-1
classifications. The author used meta-analysis to develop what he called a “criterion of safety” (CS
effect) which is a ratio expressed as follows:
No. of night accidents after lighting / No. of night accidents before lighting
CS effect ⫽
No. of day accidents after lighting / No. of day accidents before lighting (A-1)
If the ratio is less than 1.00, then it could be concluded that lighting reduces the number of nighttime
accidents. If it is greater than 1.00, then lighting increases the number of nighttime accidents. If lighting
has no effect, then the ratio would be 1.00. This is an interesting approach in that it provides a means to
prove or disprove the fundamental hypothesis. When one considers the effects of contrast on visibility, it
becomes evident that adding light to an area can diminish the contrast of a particular target and essentially
render that object invisible, which, in turn, may increase the potential for accidents. Elvik’s system can be
used to test this argument as well.
Elvik’s study found that roadway lighting reduced nighttime fatal accidents by 65 percent and nighttime
injury accidents by 30 percent. It also calculated a reduction of “property-damage-only” accidents of only
15 percent. It also found that these improvements vary by country and types of traffic environment. Elvik
recognized that the studies he reviewed did not consider every conceivable source of error. He also found
that there is “no doubt a large number of other variables with respect to which the effects of public lighting
might be expected to vary.” However, he was able to satisfy the statistical requirements for meta-analysis for
regression to the mean, secular accident trends, and contextual confounding variables. He found that the two
most significant variables were accident severity and accident type. Unfortunately, he was unable to confirm
that lighting satisfying current warrants was either more or less effective than lighting that did not satisfy
warrants. It should be noted that he found, in some cases, nighttime accident rates went up after public light-
ing was installed.
Box Study
A study of the relationship between illumination and freeway accidents (Box, 1971) concluded that the
addition of lighting reduced accidents by 40 percent. This study used a simpler ratio than Equation A-1 to
determine the effect of adding lighting:
Safety ratio unlighted = No. of night accidents/No. of day accidents (A-2)

Safety ratio lighted = No. of night accidents/No. of day accidents (A-3)
Thus the unlighted ratio is compared to the lighted ratio, and if the unlighted ratio is found to be greater
than the lighted ratio, it is concluded that lighting reduces accidents. If the reverse is true, then it is con-
cluded that lighting increases accidents.
Box concluded that freeway fixed lighting reduces accidents. It is interesting to note that his results for
Interstate 20 in Dallas show a mean ratio of only 1.01 and confidence limits of 0.72 to 1.30. In fact, the
best range in confidence limits was for Atlanta where the mean ratio was less than 1.00; that would indi-
cate that lighting increases the number of accidents. The paper also speaks to the levels of illumination
and concludes that it is impossible to determine an optimum level of illumination. Box also concludes
that those areas with the lowest illumination range had the best night/day accident ratios. This lends cre-
dence to the argument that contrast may be the salient parameter in the visibility equation.
Roadways with typical in-surface illumination levels of 0.3 to 0.6 horizontal foot-candles (HFC) had
the best accident rate ratios (Box, 1971). Great variation in luminaire output was found in the field. Data
analyzed for over 800 mercury lamps revealed wide variations in lamp output. The erratic performance
of systems invalidates any analysis of fine differences between various designs. The extent of variations
may be enough to “wash-out” meaningful analysis of small variations in lighting design. As a group,
lighted roadways had an average night/day rate ratio of 1.43 accidents of all kinds, and unlighted road-
A-2
Appendix A
ways had a ratio of 2.37. From Box’s data, a lighting level of 0.3 to 0.6 HFC produced the best ratio of
night/day accident rates. It is also interesting to note that he found that 25 percent of the urban traffic
occurs at night and the primary accident problems involve collisions due to lack of adequate acceleration
lanes. Therefore, on the issue of arbitrary target size, this study would seem to indicate a target that in
some way models the rear end of a typical vehicle. That would support a similar finding by Kahl and
Fambro of Texas A&M University (Kahl and Fambro, 1994) that the target height should exceed 150 mm.
Fisher Study
An Australian study (Fisher, 1977) went as far as to identify a point of diminishing returns with respect
to the relationship between the costs of upgrading roadway lighting systems and the savings accrued by
accident reduction. Fisher calculated a variable which he called the accident reduction factor (r):
No. of night accidents after lighting / No. of day accidents after lighting
r⫽ (A-4)
No. of night accidents after lighting / No. of day accidents before lighting
His equation is surprisingly close to the criterion of safety used by the Norwegian Elvik. He found r to
be significant at the 0.1 percent level. He also found that accident reduction was significant at the 5 percent
level with respect to lighting. This means that the change in accidents as a result of pure chance rather than
as a result of upgraded lighting could only happen in 1 instance out of 20. More important, he found that
only about 12 percent of the variation in the data can be explained by the variation in light level. Thus this
study seems to have a very sound statistical base. Fisher calculated an optimum lighting upgrade with
respect to accident cost savings. The lighting upgrade in his study was the replacement of mercury lamps
by high-pressure sodium lamps. He used a function for the lighting upgrade as expressed by the following
equation:
Lower hemisphere flux per unit area after upgrading

U⫽ (A-5)
Lower hemisphere flux per unit area before upgrading
Figure A-1. Fisher’s Optimum Lighting Upgrade Analysis (Fisher, 1977).
A-3
He compared that to a cost function (SOC), which was the savings in accident costs over increased
lighting costs due to the upgrade. Figure A-1 is a copy of the graph from Fisher’s paper and clearly shows
the optimum benefit occurs at a point around 3.3 times the increase in flux per unit area. In other words
to add more light does not amortize the additional cost of construction by a commensurate amount of
accident cost savings. Fisher also puts a very pragmatic spin on the subject of lighting and roadway safety
in the final paragraph of his paper when he states.
“Lighting does reduce night accidents and is a valuable accident counter-measure. However, there
are limits to its application, and it must be regarded as one of the many counter-measures available.
Lighting explains only a very small part of the phenomenon of accidents, and there is a diminishing
return as roadway lighting is expanded and upgraded.”
Highway Research Board Study
A similar conclusion was reached by the Highway Research Board in a study on the effects of illumi-
nation on freeways (NCHRP, 1967). They found that there was no difference in the accident rate when
illumination intensity was varied between 0.22 and 0.62 foot-candles. In fact, visibility only increased
41.1 percent with this nearly 300 percent increase in illumination because of disability glare.
Anderson et al. Study
A University of Nebraska team evaluated the impact of lighting a rural at-grade intersection (Anderson
et al., 1984) and found that the addition of lighting generally reduced accidents. However, the greatest
reduction among various designs was only about 14 percent, and in one case in the study the accident rate
actually increased 6 percent after the addition of lighting. Six different designs were studied and the vari-
ations in accident rate were less than 6 percent between differing designs. Between the two designs with
the greatest difference in accident rate, the change in average horizontal illumination was 118 percent,
which produced a 6 percent improvement in accident performance. It should be recognized that the scope
of this study was very limited, but it nevertheless shows that attempts to improve safety performance by
varying design provide only marginal differences at best.
Problem of Contrast
Taking the conclusions of the papers by Elvik, Box, Fisher, and Anderson et al. together, one can con-
clude that adding light to a road does enhance safety, and that the level of that light is hard to correlate with
safety performance. By having a lower level of illuminance, an object will show higher contrast against
both the background and the foreground when it is illuminated by the headlights of a vehicle. A paper pub-
lished in 1945 by C.I. Crouch that indicated that visual acuity rises with illumination level and then drops
off as levels of glare and brightness reach a point where the observer experiences discomfort. This identi-
fies a key biological constraint that must be considered in the design of roadway lighting systems. In
essence, we have two dichotomous conditions to try to optimize in the design. On one hand, increasing the
level of contrast makes an object more visible. This would lead an engineer to increase the light behind the
object to create a situation of negative contrast and thus maximize visibility. However, the placement of the
lighting to achieve this condition would create glare thereby reducing the observers visual acuity and
making it harder to acquire and safely react to the presence of the object in the traveled way. Additional
complications arise when off-road fixed light sources are considered along with headlamps form multiple
vehicles.
A-4
Appendix A
Jung and Titishov Study
This dilemma was addressed after a fashion by Jung and Titishov (1987). They used a standard
20 × 20 cm target, cut from a Kodak middle gray card (diffuse, 18 percent reflective standard) to conduct
their contrast experiments. They discovered the fixed lighting has too many transient quantities that are
difficult to characterize. In the case of luminance, there are only a few variables to characterize. The study
considers luminance as reflected light in the luminance design standard and illuminance design standard
as an incident light only design. It is difficult to reach agreement on standard values for visibility system
parameters when the visibility factor is loaded with physical and human factors.
Jung and Titishov’s solution is to concentrate on a less sophisticated parameter that can be computed
easily at locations on the roadway using only dimensions and properties of the lighting system. Their
parameter would be used in the same way as glare or illuminance to determine weaknesses in a roadway
lighting system. They assume visibility of a small target is determined mostly by the negative contrast of
a silhouette effect.
Jung and Titishov advocate backlighting the roadway to increase negative contrast while minimiz-
ing glare. In Jung and Titishov’s opinion, the current illuminance and luminance standards are block-
ing development of backlighting because they do not reveal spots of bad visibility. According to them,
it is necessary to perceive a critical object at a distance of about 90 m. Car headlights are not very effec-
tive at that distance, so objects are seen by silhouette vision (i.e., negative contrast) if the objects are
backlit.
Hall and Fisher Study
Hall and Fisher (1978) examined the design of roadway lighting systems by using empirically
derived requirements of light technical parameters such as road luminance, luminance uniformity, and
glare restriction. They also used a square target 200 mm × 200 mm with limited range of contrast. They
found that lighting design based on a visibility matrix requires the introduction of simplifications. They
caution that:
“Inherent simplifications may not broaden our understanding but further rigidify our [technical]
attitudes. For example, the thought that the [critical] task is the identification of simple objects on
the carriageway is reinforced. This again prevents the consideration of the total environment, which
includes the immediate surrounds of the carriageway. Indeed it may be argued that a visibility met-
ric should include a weighting factor for spatial safety distribution over the carriageway.”
These authors go as far as formally questioning the introduction of a contrast based visibility metric
because of the difficulty of understanding the effect of inherent simplifications on the output of the design
methodology.
Marsden Study
Marsden (1976) studied road lighting, visibility, and accident reduction numerically and experimen-
tally and focused to some extent on the issue of glare. For experimental investigation, disability glare is
related to veiling luminance, which was measured with a Pritchard photometer. Horizontal illuminance
near the road surface was measured by summing the outputs of photocells mounted on the ends of the
vehicle. Vertical illuminance at road level was measured by a photocell mounted on the rear of the vehi-
cle, and some instrumentation was mounted below the vehicle to record road reflectance data. They
recorded all the information as well as the visual field of the driver on the tape. The tape was played in
the laboratory and selected frames were frozen. An area of the shape can be defined (by operating bright-
ening-up controls) for luminance analysis. This analysis was examined on the portion of the TV signal
A-5
corresponding to the selected area. Analog processing gives the value of maximum, minimum, average,
and standard deviation of luminance within the selected area by using a calibration luminance scale on
the picture.
Scott Study
In Scott’s study (1980) a wide range of measures of lighting quality was investigated in order to deter-
mine which of them was most clearly related to accident frequency. A representative number of sites was
examined to establish correlations with reasonable confidence. It was assumed that day and night acci-
dent frequencies would be equally distributed in all factors except road lighting quality.
Analysis was limited for dry road surfaces and major two lane roads with speeds greater than 30 mph.
It was concluded that average luminance is strongly related to accident ratio. An increase of 1 cd/m2
resulted in a 35 percent lower accident ratio. A non-pedestrian accident ratio correlated well with average
luminance and overall uniformity (U0). A significant relationship with average luminance was found for
minor accidents.
Frantzeskakis Study
J.M. Frantzeskakis (1983) presented the results of two accident analysis studies conducted on two
highway sections of the Greek National Roadway Network. It was found that overall accident rates
increase during winter, on weekends, at night, and during the period before sunset. It is considered more
appropriate to correlate accident rates to levels of service instead of traffic volumes.
A.2 DRIVER PARAMETERS
Introduction
Rackoff and Rackwell (1975) investigated the physiological components of driver reaction and target
acquisition. They developed a vehicle-based television system to investigate driver eye movement pattern
during night driving and to compare those patterns to daytime patterns on freeways and a rural highway.
They determined the differences of visual search behavior at sites with high and low night accident rates
and the effect of illumination on a driver’s visual search. They discovered that nighttime visual search
behavior is different from daytime visual search behavior, and the measure of visual search behavior is
sensitive at sites with different day and night accident rates. The results demonstrate that the changes in
visual search measures due to illumination not only demonstrate that illumination can affect visual search
at the same sites, but also demonstrate that visual search behavior can be useful in associating the specific
effects of various illumination designs on driver search patterns.
Walton and Messer (1974) discuss fixed roadway lighting from a driver visual workload measure of
effectiveness of vehicle control. They were looking for a measure for determining when roadway light-
ing would be warranted. Their work compliments the concept discussed earlier with regard to target
acquisition time, reaction time, and stopping time.
Driving Tasks
Walton and Messer divide driving into three primary tasks, the information necessary to complete
each task, and the priority level of each task. The tasks and priority levels are the positional level, the
A-6
Appendix A
situational level, and the navigational level, respectively. The positional level consists of speed and
lane position and must be satisfied before any other task. The situational level is second and consists
of changing speed, direction of travel, and position on the roadway. The navigational level consists of
following a predetermined route from here to there and is the third level of priority after position and
situation.
In a situation overload, a driver will shed lower priority tasks for high priority tasks. An environ-
mental situation causing a driver to shed high priority tasks is not a suitable situation. Load shedding
is not determined by the amount of work a driver must do but by the rate at which the tasks must be
accomplished. An emergency situation will cause sudden load shedding. From an information supply
standpoint, the size of the information supply to the driver is inversely proportional to the speed at
which he is traveling. Fixed roadway lighting improves information processing capability of drivers by
increasing the amount of information available for processing by making a larger proportion of the
roadway visible.
To quantify the amount of information available due to fixed lighting, we first need to determine the
total amount of information available to the driver under ideal lighted (i.e., daytime) conditions. Then, we
must determine the amount of information available in the same area at night without lighting, which then
allows the computation of the contribution of the fixed lighting in terms of total information available to
a driver. After the information contribution due to fixed lighting is assessed, it is then possible to deter-
mine the change in information available to a driver due to changes in fixed lighting.
Drivers are assumed to service information needs in a cyclic order dictated by priority of tasks. The
cycle would be positional information search, situational information search, navigation information
search and back to positional information search. From an information standpoint, the tasks involve sam-
pling each task periodically with the period of the sample determined by the speed of the vehicle and
complexity of the task. As a task becomes more complex, the sample rate increases.
The assumption of safe and effective vehicle positional control is based on redundant positional infor-
mation of the roadway ahead and must be acquired each time the driver returns to a position information
search and acquisition phase. During situational information search and navigation information search,
the driver is assumed to be traveling without positional information. Information demand is the time
required to complete a sequence of position, situation, navigation, and position information searches.
Positional Information Acquisition
Most night time positional information is gathered from lane lines, edge lines, curb lies, position of
other vehicles, and a general view of the roadway. Much of the positional information under good (day-
light) driving conditions can be obtained with peripheral vision. During nighttime driving, the driver
fixates on position markers rather than using peripheral vision. Time required to identify a task is about
0.2 seconds. The time for eye movement is from 0.1 to 0.3 seconds. So, the time required to sample a
position source is about 0.3 seconds or more.
Situational Information Acquisition
It is assumed that a driver scans situational areas to ensure safe operation when a potential hazard is
visible about 25 percent of the time, but if there are no hazards, the situational load drops. Increased
complexity of the scene being viewed increases the mean fixation time of the situational information
gathering tasks.
Navigational Information Acquisition
A driver can search for navigational information only after the positional and situational needs are ful-
A-7
filled. Navigational information consists of reading signs and other navigation tasks. The complexity of
the tasks is determined by a level of familiarity with the route and with the situation. New signs and situ-
ations require more time and increase stress levels. A word on a sign requires about 0.35 seconds to locate
and read. Multiple unfamiliar signs are confusing and increase stress levels during navigation tasks. As
navigational task time increases, positional and situational task times suffer. Roadway lighting increases
the positional information supply by increasing the visibility distance. Decreasing speed also increases
visibility distance.
Addressing Drivers’ Information Needs
Walton and Messer’s approach to warrant fixed roadway lighting is based on the driver’s information
needs to perform night driving tasks in a particular driving environment. Fixed roadway lighting is war-
ranted when the information demand exceeds the information supply without fixed roadway lighting.
Adrian (1997) adds to the knowledge base with respect to driver physiology. He discusses rod vision
and cone vision and the 2 degree central field of view and blue shift in the eyes sensitivity. He also found
that as light levels decrease, the spectral sensitivity of the eye changes, but the sensitivity curve remains
approximately the same shape. However, the peak of the curve shifts away from 550 nanometer (nm), to
a slightly bluer 520 to 530 nm. Low light level contrast sensitivity is shifted into the blue with higher con-
trast sensitivity in blue than in red.
Target Size and Composition
The proposed RP-8 (IESNA, 1990) specified that size and composition of the “small target” to be
18 cm square and of 20 percent diffuse reflectance (change to 50 percent before adoption in 2000). This
reference is silent as to the reasons why this particular target is chosen as the standard. It is clearly an
attempt to create a series of parameters that can be related to visibility and, therefore, correlated to exper-
imental and computed data with for the purpose of quantifying visibility. A study led by Freedman
(Freedman et al., 1993) proved that the probability of detecting a target strongly depends on its type and
that older driver’s generally showed a significantly lower probability of target detection. Thus, the selec-
tion of a target’s size, shape, and composition should not be arbitrary. Other studies have used targets of
different size than the STV target (the term target will be used to define a standard object used experi-
mentally to relate to some other parameter of visibility, recognition, or other such factor). Roper (1953)
used targets 40.64 cm square with a reflectance of 7.5 percent. Haber (1955) used a much larger target
with a mean linear dimension of 91.4 cm and a reflectance of 15 percent. A German group (Waetjen et al.,
1993) used a target composed of a Landholt ring with a stroke width of 8.7 cm and a height of 43.5 cm.
Jung and Titishov (1987) conducted their work with a 20 cm square target with a reflectance of 18 per-
cent. Zwahlen and Schnell (1994) used targets of varying reflectances that were 60.96 cm square and
installed 30.48 cm above the pavement. They did further detailed studies on this type of target with a con-
stant reflectance of 15.5 percent.
A team led by Janoff (Janoff et al., 1986) used a target composed of styrofoam hemisphere with a
0.15-m diameter skirt and an 18 percent reflectance. The lighting system in controlled field conditions
consisted of 200-watt, high-pressure sodium (HPS) lamps mounted 30 ft high at spacings of between 68
and 88 ft. They chose six different lighting conditions: full lighting, 75 percent power, 50 percent power,
every other luminaire extinguished, one side extinguished, and no lighting. They measured photometric
data for each condition. Subjects were required to drive the vehicle at the 55 miles per hour (88 kph) con-
stant speed limit. Results of the controlled field experiments show that drivers tended to dislike reduced
lighting on ramps or interchanges as opposed to reduced lighting on straight mainline roadway sections.
By using the six conditions, they found a linear relationship between detection distance and horizontal
illumination, and between pavement luminance and visibility index
Zwahlen and Yu (1990) studied two types of investigations to determine the distances at which the
A-8
Appendix A
color and outside shape of targets can be identified at night under vehicle low-beam illumination for flat
targets with three different outside shapes and with six different retroreflective color sheet coverings.
First, the color and the shape recognition distances were investigated. Second, only the color recognition
distance was determined. They used colors (red, green, yellow, orange, blue, and white) and target shapes
(circle, square, and diamond) having the same surface area (36 in2) as independent variables. In both
experiments the center, front of the vehicle is positioned above the centerline of the road, and the longi-
tudinal centerline of the vehicle also positioned at a 3 degree angle to the left of the road centerline. The
results show that the color recognition distance was twice as long as the shape recognition distance. Also,
they concluded that a highly saturated red color of the retroreflective targets was the best. Hall and Fisher
(1978) examined design of roadway lighting system by using empirically derived requirements of light
technical parameters such as road luminance, luminance uniformity, and glare restriction. They used a
200-mm square target with limited range of contrast. They found that lighting design based on visibility
matrix gives better results than others. Finally, the 1990 “Green Book” (AASHTO, 1990) uses a target
150 mm in height as a standard from which to calculate stopping sight distance requirements for highway
geometric curves. Thus, it can be seen that target size and composition has been quite variable.
While roadway lighting can be installed for a variety of purposes, the consensus found in the literature
seems to indicate that safety is the primary reason for making a capital investment in lighting systems.
Thus, it would seem logical that the size and composition of the standard target used for design would
be directly related to the dynamics of nighttime driving safety. A study done by Kahl and Fambro (1994)
provides an excellent analysis of the comparison of targets to accidents. This pair correlated types of acci-
dents with the size of the object involved and then compared it to the standard Green Book 150-mm tar-
get. They found that only 0.07 percent of reportable accidents were attributable to collisions with small
objects in the road. They then concluded that the frequency and severity of these types of accidents did
not justify the use of the 150-mm object height in the critical Stopping Sight Distance model. In fact, they
found that only 2 percent of all accidents involved objects or animals in the roadway. In urban areas, 10.4
percent of the objects struck were less than 150 mm in height, and on rural roads only 1.8 percent were
150 mm or less in height. They also found that “more than 95 percent of the accidents resulted in low-
severity injuries; therefore, a small object is not the most critical, hazardous encounter in the Stopping
Sight Distance situation.” They also make two recommendations of interest in the STV discussion:
• The object height should be a function of and related to the smallest realistic hazard typically
encountered on the roadway.
• The taillight height of an average vehicle (380 mm) is probably a good measure for the height of a
typical hazard.
This correlates well with the results of Zwahlen and Schnell (1994) who found that a 60.96-cm square
target with 15.5 percent reflectance placed at 30.48 cm above the pavement could be spotted by subjects
at an average distance of 104 m with a standard deviation of 16.6 m through the filter of a windshield.
When this is compared to the STV model of the 18-cm target visible at 83 m, there appears to be a poten-
tial that the STV target might be too small to be detected by the average observer, and that the use of it
as a design standard does not directly correspond to those hazards related to visibility for which the
lighting is being installed. The Zwahlen and Schnell target provides nearly three times the reflective sur-
face at nearly the same distance (if one were to subtract the standard deviation from the mean distance)
as STV. It should be noted that the Zwahlen and Schnell experiment was a static one in that the observer
was not moving as would normally be the case in most roadway hazard situations. Also, the observer’s
only data collection task was to search for the target. Jung and Titshov (1987), while using a target that
was very close to that specified by STV, found that once “luminance levels meet standards for unifor-
mity, spots of unsafe low contrast are clearly revealed . . .” They also seem to advocate the use of sev-
eral standard values of reflectivity.
A-9
A.3 PAVEMENT PARAMETERS
Background on Pavement Luminance in Roadway Lighting Design
Until the 1983 IES/ANSI Standard Practice for Roadway Lighting (RP-8) was proposed, roadway
lighting in North America was based on horizontal illuminance. In the 1983 RP-8, pavement luminance
was introduced as the preferred basis of design with illuminance criteria included as an acceptable alter-
native (IES 1983). High mast and walkway-bikeway lighting systems were two exceptions where illumi-
nance was presented as the only criterion for design.
Along with pavement luminance, disability glare (veiling luminance) was also identified as a signifi-
cant factor that affects the nighttime visual performance of a driver. At the time when RP-8 (IES 1983)
was introduced, the IES/ANSI recognized that “luminance criteria do not comprise a direct measure of
the visibility of features of traffic routes such as traffic and fixed hazards.” However, they decided that
“visibility” criteria proposed at the time were based on limited research and evaluation and, therefore,
could not be adopted at that time. Nevertheless, RP-8 (1983) had a complete appendix (Appendix D) ded-
icated to visibility concepts for information purposes. This Appendix used the concept of visibility index
(VI), which was developed based on research by Blackwell and Blackwell (1977) and Gallagher (1976)
where the visibility of a gray-colored rubber traffic cone was considered as the target.
According to literature, the “visibility” concept was first introduced in England by Waldram (1938)
who identified the concept of “revealing power.” He calculated the visibility for 24-inch square targets set
on a grid pattern on the roadway, and determined when a target became a dangerous obstacle for
observers driving at 30 mph. Also in England, Smith (1938) conducted a study of the reflectance factors
and revealing power of objects. He showed that 50 percent of the pedestrian clothing had a reflectance of
less than 5 percent, and 80 percent of the clothing had a reflectance below 15 percent. Based on these
results, it was possible to show that a 10 percent reflectance target will always be darker than the pave-
ment that can act as a background for a pedestrian wearing such clothes. Such a scenario provides nega-
tive contrast (pavement brighter than the target), and the target that is least visible on the roadway will be
the one located where the pavement has the lowest luminance.
Based on research such as those mentioned above, CIE (International Commission on Illumination)
adopted the following positions in its standard practice for roadway lighting design:
• “Quality” of a lighting system is always higher when average pavement luminance is high.
• “Quality” of a lighting system is always higher when “empty street” pavement luminance uniformity
is excellent.
• Glare needs to be considered in the design.
It was interesting to note that contrast was not considered as a design criterion. Keck (1996) observed
that CIE at the time felt that objects are almost always darker than the pavement, and therefore, by con-
sidering factors 1 and 2 above, would provide a simple design method.
Reflective Properties
In terms of reflective properties, all surfaces, including roadway pavement surfaces, are generally clas-
sified into three major groups. These are the ideally specular surface, perfectly diffuse surface, and
mixed-reflection surface. Descriptions of each follow.
Ideally Specular Surface. The ideally specular surface reflects all the luminous flux received by a
point at an angle of reflection equal to the angle of incidence and in the same plane. The reflected ray, nor-
mal to the surface at the point of incidence and the reflected ray all lie in the same plane. These surfaces
form a similar geometric image. Some examples of almost ideally specular surfaces are mirrors, highly
polished metal surfaces, and liquid surfaces.
A-10
Appendix A
Perfectly Diffuse Surface. The perfectly diffuse (matte) surface reflects light as a cosine function of
the angle from the normal, regardless of the angle of incidence. Since the luminance of a surface is equal
to intensity divided by the projected area, and since the projected area is also a cosine function of the
angle from the normal, the perfectly diffuse surface appears equally bright to an observer from any view-
ing angle. The luminance of the surface is independent of the luminance of the source of light but pro-
portional to the illumination of the surface. These surfaces form no geometric image. Surfaces such as
white matte finished paper or white painted walls closely approximate the perfectly diffuse surface.
However, these surfaces behave as diffuse only if the angle of incidence is close to zero.
Mixed-Reflection Surface. Most surfaces encountered in everyday life fall into the category of mixed
reflection that is somewhere between the ideally specular and perfectly diffuse surfaces. These surfaces
form no geometric image but act as a diffuse surface to some extent with some preference to direction of
reflection. Therefore, the apparent brightness of such a surface changes with the angle of incidence and
the observer’s viewing angle. King (1976) illustrated these surfaces with the luminous intensity distribu-
tion curves shown in Figure A-2.
Pavement surfaces that encounter viewing angles between 86 and 89 degrees and incident angles
between 0 and 87 degrees (both from the normal) exhibit characteristics of mixed reflection. Generally, a
single luminaire over the pavement surface produces a single luminous patch that appears to the traveler
to be shaped like a “T” on the surface of the roadway with the tail of the “T” always extending towards
the observer irrespective of the observer’s position on the roadway. This brightness (luminous) patch is
almost completely on the observer side of the luminaire since very little of the light incident in the direc-
tion away from the surface is reflected back to the observer. The size, shape, and the luminance of the
patch depends to a great extent on the surface reflection characteristics of the pavement.
Figure A-3 illustrates the shape of a luminous patch produced by a luminaire on diffuse, smooth (spec-
ular), and wet surfaces.
Figure A-2. Luminous intensity distribution curves for different types of reflection (King
1976).
A-11
Figure A-3. Luminous patch produced on different roadway surfaces (King 1976).
Studies on Pavement Reflectance
In one of the earliest studies done on reflection characteristics of pavement surfaces, Christie (1954)
of the Transportation and Road Research Laboratory (TRRL) in England found that a reduced range of
data presented in a single family of curves is sufficient to calculate the luminance in important regions of
a street lighting installation with an accuracy level of 15 percent. Christie adopted the view that bright-
ness (luminance) in a lighting installation is built up from the bright patches produced by individual lumi-
naires. This technique was used to assess the reflection characteristics of three types of pavement surfaces
A-12
Appendix A
commonly used in England. The three surfaces were rolled asphalt with precoated chippings, “non-skid”
rock asphalt, and machine finished Portland cement concrete. After calculating the luminance factors, a
family of curves for these surfaces was drawn. Since these curves do not present an immediate picture of
how the surface is brightened, a perspective drawing showing the bright patch was developed using the
luminance factor curves.
Comparing his brightness patch for the rolled asphalt surface with precoated chippings with results
published by Waldram (1934), Christie concluded that the old surface (Waldram’s test section) gave a
much larger brightness patch than the new surface (Christie’s test section). In comparing the rolled
asphalt surface with the “non-skid” rock asphalt, Christie observed that the efforts to make pavement sur-
faces “non-skid” have seriously reduced their power to reflect light. Christie also found that in addition
to coarse surfaces, fine textured surfaces with protruding small aggregates also produce short brightness
patches. His explanation of this phenomenon was that within limits, what matters is the shape of the sur-
face, not the size of its features (coarseness). He concluded that sharp projections necessary to prevent
skidding tend to destroy the specular reflection of obliquely incident light that makes possible the forma-
tion of long patches. Christie also observed that in coarser surfaces where specular reflection is reduced,
brightness has to depend more on diffuse reflectance than in the case of smoother surfaces. Since diffuse
reflectance depends on the lightness of color, he said that the benefits of using light-colored materials
should be substantial. Christie also commented on how the type of luminaire can be changed to overcome
problems involving smaller brightness patches. On skid resistant coarse surfaces, he suggested that high-
angle beam luminaires are not very satisfactory and medium angle luminaires with maximum intensity at
75 to 78 degrees are preferred.
Finch and King (1967) appear to have introduced the first direct reading reflectometer for roadway
lighting purposes. Until then, reflective characteristics of pavements were evaluated using visual pho-
tometry and other photographic techniques. This device allowed full flexibility in changing all three
angles relating to reflectivity. It operated on 115-volt AC power, and it used a stray light rejection curve
for the telephotometer where the light acceptance angle was approximately three minutes. The problem
associated with this device is that it took approximately three hours to set up the equipment at site and
another hour to take one set of reflection data corresponding to a set of angles. If measurements were
taken at 5 degree intervals for the vertical source inclination and the horizontal angle, it would result in
864 readings at one location and require 864 hours of data collection. King and Finch (1968) later devel-
oped a reflectometer for use in the laboratory where 12-in. diameter pavement cores were used to simu-
late the pavement. By automating the data collection procedure, they were able to make rapid automatic
readings of the directional reflectance factor, thus enabling the collection of large volumes of data over a
very short time. This device was able to simulate up to 600 ft of viewing distance. One significant feature
of this reflectometer was that the color response was corrected to approximate that of the human eye.
Even after the development of their automated pavement reflectance measurement device, King and
Finch (1968) observed that there was little application of it outside the research laboratory, primarily due
to the specialized nature and complexity of calculations involved. They suggested that one way to expand
the use of reflectometry is to use a pavement surface classification system and proposed that the classifi-
cation be based on directional reflectance properties of the pavement surface.
Towards the latter part of the 1970s, the University of Toronto (Jung et al., 1984) built a photometer
for the road surface reflectance measurement based on concepts developed earlier by CIE (1976). This
reflectometer features automated control of positioning and reading and recording data. It is capable of
testing pavement cores 6 to 8 in. in diameter and at least three core samples from a given pavement are
required to classify the pavement type. Jung et al. (1984) conducted a study to measure reflectance prop-
erties of many types of pavements in Ontario. The measurements were made on 6-in. diameter cores
taken from 36 different pavements where more than 400 core samples were processed. When factors such
as traffic level and the position of the lane were considered, this accounted for about 100 different surface
types.
Pavement surfaces were classified based on the average luminance coefficient Q0, and the ratios S1 and
S2 as defined by IES Roadway Lighting Committee (1976). Q0 is considered as a measure of the overall
A-13
brightness of the pavement as it appears to the viewer, whereas S1 and S2 describe the degree of specular-
ity of the pavement surface. Over the years, two systems of four standard reflectance tables have been
proposed for dry pavements. These two systems are indicated by “R-Series” and “N-Series” classifica-
tions. This guide adopts the R-Series classification for its pavements and its features are indicated in
Table A-1.
Theoretical Basis for “R-Series” Classification of Pavement Surfaces
The classification is based on specularity of the pavement as determined by a ratio, S1, and a scaling
factor, Q0, as determined by the overall “lightness” of the pavement. The normalized Q0 is given in table
18.2 for each of the pavements described. Greater accuracy in predicting pavement luminance can be
achieved by evaluating specific pavements as to their S1 ratio and specific Q0, and then choosing the cor-
rect r-table.
The S1 ratio and specific Q0 for a pavement can be determined in one of two ways:
• A core sample can be removed from the pavement and photometered by a qualified laboratory.
• A field evaluation can be made.
The characteristics S1 and Q0 were adopted by CIE as basic quantities for evaluation of the reflection
properties of a road surface (CIE Publication 30, 1976). The average luminance coefficient (scaling fac-
tor) Q0 is given by the following equation as a measure of the lightness of a road sample:
1 ⍀0
Q0 ⫽
⍀0 ∫0
q( g, b)d⍀ (A-6)
In the formula 18-6, W0 is the relevant solid angle of incident light at a specified point on the road.
W0 is defined by a rectangular “ceiling” at the mounting height h extending 3h to the right and left to
the specified point, 4h toward the observer and 12h behind the specified point. The special quantity S1
given by:
1 ⍀0
Q0 ⫽
⍀0 ∫0
q( g, b)d⍀ (A-7)
Figure A-4. Schematic diagram of roadway lighting.
A-14
Appendix A
The angles g and b are as shown in Figure A-4.

This function is derived from the angular distribution of the reduced luminance coefficient to indicate
the shape of the reflection indicatrix:
r ( g, b) ⫽ q( g, b)cos3 g (A-8)
–
It was found that Q0 is highly correlated with the average luminance L on the road as follows:
1 A
L⫽
A ∫
0
Lp dA (A-9)
where Lp is the luminance at point P and A is the relevant portion of the road area (usually restricted to
one luminaire spacing).
Based on calculations of 24 road surfaces, 24 luminous intensity distributions, and 72 one-sided
lantern arrangements, the correlation coefficient between reflection characteristic Q0 and average lumi-
–
nance L was found to be 0.96 (Schmidt, 1986). Bodmann and Schmidt (1986) conducted field experi-
ments to compare calculated and measured luminance characteristics. The reflection characteristics, Q0,
were measured with the LTL 200 portable road surface reflectometer.
–
The average luminance L was measured with the portable luminance meter. On average, the calculated
–
values for L were found to be 31 percent higher than the measured values. The experimenters estimated
the results as a reasonable estimate. If one takes into account maintenance factors such as the decrease of
light output with age and deterioration of reflecting and transmitting materials of the luminaires, the
agreement between calculated and measured luminance can be evaluated as perfect.
TABLE A-1. Reflectance Parameter Values for “R” and “N” Classifications (Jung et al., 1984)
R-Series N-Series
Parameter R1 R2 R3 R4 N1 N2 N3 N4
Q0 0.10 0.07 0.07 0.08 0.10 0.07 0.07 0.08
S1 0.25 0.58 1.11 1.55 0.18 0.41 0.88 1.61
S2 1.53 1.80 2.38 3.03 1.30 1.48 1.98 2.84
Table A-1 identifies the values for the three reflectance parameters Q0, S1, and S2. The N-Series clas-
sification was developed in Germany by Erbay (1974). One note of caution by CIE was that variation of
Q0 within one pavement class might be very high, so the Q0 value given within the standard must be
scaled to correspond to the Q0 of the actual surface being chosen.
TABLE A-2. Relationship between Coarse Aggregate on Pavement Surface and S1

(Jung et al., 1984)
Coarse Aggregate Range of Log (S1)

Igneous or Trap Rock –0.29 to –0.17
Limestone –0.10 to –0.06
Blend of the Above Two Aggregates –0.23 to –0.08
A-15
TABLE A-3. Relationship between Coarse Aggregate on Pavement Surface and Q0

(Jung et al., 1984)
Coarse Aggregate Range of Log (Q0)

Dark Trap Rock 0.074 to 0.088
Bright Limestone 0.102 to 0.124
Blend of the Above Two Aggregates 0.086 to 0.097
Pavement Reflectance Studies
Table A-2 outlines the range of Log (S1) values obtained for three different types of coarse aggregates.
Jung et al. (1984) attributed the different values to different levels of resistance to polishing under traffic.
This data quite clearly indicated a relationship between the S1 ratio and the type of coarse aggregate used
on the pavement surface.
As for Q0, a wide scattering of values was observed. Table A-3 outlines the range of Q0 values obtained
for three different types of coarse aggregates. Jung et al. (1984) attributed the different values to different
brightness levels of aggregate and on a concurrent increase in specularity of the surface. The appropriate
value to be used in design depends on the road surface materials, their composition and changes to the
pavement surface with time and traffic exposure. Jung et al. (1984) also noted that with time, asphalt
pavements tend to brighten and Portland cement concrete tends to darken. They also observed that with
coarse aggregates that are polishable due to traffic, there might be a shift in the specularity class, for
instance from R2 to R3.
Some notable observations made by Jung et al., (1984) included changes in specularity between dif-
ferent lane wheelpaths and that asphalt pavements become more specular as reflected with increased
values for Q0, S1, and S2. In the end, it was reported that with regard to specularity only, the four pave-
ment classes were regarded as sufficiently accurate for design purposes. However, the authors cau-
tioned that, due to high variability observed, Q0 should be estimated more accurately by considering
the surface course composition and the aggregate type. Nevertheless, based on their extensive mea-
surements of pavement reflectivity, the authors published recommended (and amended) design values
for different combinations of coarse aggregate type and mix design commonly used in Ontario,
Canada.
Bodmann and Schmidt (1989) showed the marked variation in the reflection characteristics of road
surfaces with time and traffic and highlighted some problems associated with the CIE recommended stan-
dard classes of pavement surfaces. They also pointed out that the decision on the class of surface to be
used in the design is often based on assumptions and the standard R-table represents the individual road
surface irrespective of temporal and local variations due to age and wear. Furthermore, the authors indi-
cated that the classification of surfaces into four CIE “R” classes is justified neither by test calculations
nor by measurements on real streets. Based on these observations, the authors highlighted the positive
aspects of the “C-Series” classification where only two standard surfaces are considered. The two classes
of pavement surfaces are C1 and C2, C1 corresponds to the R1, and C2 corresponds to R2, R3, and R4 in
the “R-Series” classification. The authors contend that the “C-Series “ classification for dry road surfaces
is more realistic and much more practicable. However, even under the “C” classification, the prediction
of Q0 remains a problem at the design stage.
Nielson et al. (1979) studied the reflectance characteristics of 41 different road samples, 24 of which
were asphalt concrete, and the rest were hot-rolled asphalt with coated chippings. The surface materials
were cast into 30 × 35 cm rectangular specimens and were tested in the laboratory. In addition to the mix
type indicated above, the maximum size of the aggregate, aggregate type, and the climatic conditions
were included in the experimental design. The results from this study can be summarized as in
Table A-4.
A-16
Appendix A
TABLE A-4. Summary of Observations by Nielsen et al. (1979)
Factor(s) Investigated Conclusions and Observations
Parameter Relationships Q0 and S1 are inversely correlated.

Retroreflection QR does not change significantly with the observation distance.
Specular reflection QS on dry roads increases with meeting distance to
oncoming vehicle.
Specular reflection QS increases from dry to wet pavement by a factor of two.
For worn samples, Q0 and QR are directly correlated.
Surface Wear The brightness measured by Q0 and QR develop slowly with wear.
Residual bitumen (after aggregate is polished) increases S1.
Reflection properties increased slightly when summer tires are used, but
improved during subsequent exposure to studded tires.
Composition Brightness of aggregate affects Q0 and QR.
Brightness not affected by coarseness of aggregate.
According to Bodmann and Schmidt (1986), if a decision has to be made as to whether or not a partic-
–
ular road lighting installation meets prescribed values, the tolerance (1 ± 0.1)L can be recommended.
Analysis of computer predicted luminances was conducted by (Janoff, 1993). The illuminance calcu-
lations performed in the past using computers were quite accurate when given all input parameters.
However, in 1983 the roadway lighting standard changed based primarily on pavement luminance
(Janoff, 1993). This change brought about more complex calculations. It also required the exact
reflectance properties of the pavement surface. This made the computer programs accuracy very depen-
dent on factors such as the r-tables, the formulae for computing pavement luminance, lighting geometry,
and the luminaire intensity distribution (Janoff, 1993). A standard practice based on visibility was pro-
posed in 1990 (Janoff, 1993). The visibility level (VL) can be determined using photometers to measure
target luminance, pavement luminance, and veiling luminance. A predictive computer program thus
becomes an important tool for deriving target, pavement, and veiling luminances for a road lighting
design in progress. In 1992, the only predictive computer program available was STV. A study was per-
formed to compare the target, pavement, and veiling luminances, as well as VLs, to measured values. This
experiment consists of two different targets. Each target was a 7-in. square, with one placed upstream of
the closest luminaire and one downstream (Janoff, 1993). The targets consisted of three different
reflectances: 5, 30, 80 percent. During this study there were 48 measured points. For accurate measure-
ments all street lights were cleaned, aligned, and 12 new calibrated lamps were installed closest to the tar-
get locations. The results indicated that the predicted values did not match up with the measured values.
There were significant differences between the target, pavement, and veiling luminances (Janoff, 1993).
For example, during one experiment the veiling luminance (Lv) was measured and predicted at 275 ft for
each target. The results are shown in Table A-5.
TABLE A-5. Veiling Luminance
Variable Target Reflectance Target Position Measured Predicted Result Ratio
Lv — 1 2.10 0.16 13.1

Lv — 2 2.86 0.22 13.0
A-17
The measured value at target position 1 was 2.10 and the predicted value was 0.16. It was concluded
that “. . . many of the problems may not be in the computational parts of STV, but rather in the (inaccu-
rate) choice of r-tables, choice of nominal target reflectance, specification of proper candlepower distrib-
ution, or failure to include light reflected from pavement onto target.” (Janoff, 1993).
An effect of headlights on luminance and visibility was studied experimentally. There are at least two
relevant parameters to consider in roadway visibility. The first is headlight intensity in the direction of
the road ahead, and the other is the intensity in the direction of the eyes of the driver (Alferdinck et al.
1988). A study was performed to evaluate the increase of visibility due to the addition of vehicle light-
ing. A number of measurements were made under 20 different lighting conditions to determine increases
in visibility. The reported measurements taken used 5, 30, and 80 percent target reflectances. The mea-
surements were taken with and without headlighting. Measurements were taken first at 75 ft then every
50 ft up to 275 ft; then they were taken every 100 ft up to 775 ft. The study concluded that at distances
less than 275 ft, there is a significant change in photometric visibility resulting from headlights (Janoff,
1992). However, at distances greater than 275 ft there is no effect of headlights on either small target vis-
ibility or recognition distances derived from subjective estimates provided by the drivers (Janoff, 1992).
Adrian et al. (Unpublished Data) studied the influence of light reflecting from the road surface on to
the target on STV. Their results showed that this indirect portion of illuminance can contribute up to 15
percent of the total target illuminance. This will significantly alter the VL required to see the target under
positive contrast.
A-18
A.4. REFERENCE SUMMARY
RP-8 Assumption/
No. Ref. Reference Abstract Limitation Main Conclusions
1 1 Uniformity of Illumination in A review of design procedures adopted by Glare not considered. When recommended procedures such as
Highway Lighting. D. M. many highway departments and local the average illumination are used, wide
Finch and A. E. Simmons government agencies are analyzed in variations in roadway illumination and
(1950), Journal of the IES 45, terms of parameters associated with good brightness pattern may result even though
p. 561, Univ. of California, lighting design. A simple method was the average illumination is maintained.
Berkeley, California. developed whereby the designer could Minimum illumination criterion should
quickly get all essential design be used rather than the average.
information.
2 3 Comparison of Accidents and The economic impact on the city of Street light glare and Street classification appears to be more
Illumination. Paul C. Box, Syracuse for upgrading street lighting to ambient light not significant in accident-illumination
(1972), HRR 416, pp. 1–9, national standards is determined. The considered. relations than the abutting land use
Syracuse, NY. methodology developed during this classification.
research may be the state-of-the-art for Streets with high-illumination levels
similar future studies. tended to have above-average night/day
accident ratios and accident cost ratios.
3 5 The Relationship Between An investigation of a wide range of Day/night accident Average luminance is strongly related to
Road Lighting Quality and measures of lighting quality was frequencies would accident ratio.
Accident Frequency. P. P. conducted to determine which were most equally reflect between An increase of 1 cd/m2 result in a 35
Scott, (1980), TRRL clearly related to accident frequency. site variations in all percent lower accident ratio.
Laboratory Report no. 929. Also, a sufficient number of sites were factors other than road A non-pedestrian accident ratio correlated
London, U.K. selected to establish relationships with lighting quality. well with average luminance and overall
reasonable high confidence. Analysis is limited to uniformity.
dry road surfaces and A significant relationship with average
major two lane roads luminance was found for minor accidents.
with speeds .30 mph.
4 7 The Specification of Street Objectives of the study were to:
Lighting Needs. V. Gallagher, • Develop a visibility metric for
B. Koth, and M Freedman roadway lighting design,
(1975). FHWA Rep. RD-76-17 • Determine the precision of this metric
Franklin Institute, Philadelphia,
Pennsylvania. as a measure of driver visual
performance,
• Develop spec. for driver visibility needs,
A-19
Appendix A
• Develop a method to predict visibility

with lighting geometry, luminaire
specs and pavement type
RP-8 Assumption/
A-20
5 8 Effectiveness of Highway Cost-benefit analyses of arterial highway Total nighttime dry-weather accidents are
Arterial Lighting. M. S. lighting treatments were conducted with inversely proportional to visibility.
Janoff, B. Koth, W. McCunney, regard to traffic safety and energy usage. HPS luminaires are more cost beneficial
M. Freedman, C. Durek, and Regression equations were developed to than mercury luminaires. However, both
M. Berkovitz (1977), FHWA predict nighttime accident history based types provide a cost/benefit ratio greater
Report RD-77-37 on population density, area classification than 1.
Franklin Institute, Philadelphia, and visibility. When visibility and accident reduction are
Pennsylvania. A design guide was prepared to assist in constraints, 400W HPS luminaires gave
conducting cost-benefit analysis of optimum designs.
lighting improvements. When cost and energy use are constraints,
150W HPS luminaires gave optimum
designs.
6 9 Cost–Effectiveness There are no generally accepted design At each wattage level, traffic operations
Evaluation of Rural criteria that define the levels of were safer under a two-luminaire system
Intersection Levels of illumination required at rural than they were under a one-luminaire
Illumination. Kyle A. intersections. The objective was to system.
Anderson, Weldon J. Hoppe, evaluate the cost-effectiveness at different Two 200-watt HPS had the lowest total
Patric T.McCoy, Ramon levels of illumination. annual cost at intersections with main
E. Price, (1984), TRR 996, highway ADT >3750.
pp. 44–47. Omaha, Nebraska. No lighting system was warranted at ADT
< 3250.
7 10 Freeway Lighting and Traffic A case study on traffic accident The savings in energy Accident rates were reduced due to
Safety—A Long-Term characteristics was conducted that utilized costs after switching lighting and increased after lights were
Investigation. R. Lamm, J. M. a suburban freeway area in Germany to off lights as compared switched off at night to save energy
Kloeckner, and E. M. Choueiri, assess the effectiveness of freeway with savings in (Partial Lighting).
(1985), TRR 1027, pp. 57–63. lighting. accident costs was not
Frankfurt, Federal Republic of determined (it deals
Germany. only with safety).
8 11 Effects of Turning Off To save energy, continuous roadway Ramp and frontage The savings in power consumption
Selected Roadway Lighting as lighting were turned off and two years of road lighting were associated with a cut back are offset to a
an Energy Conservation before-data and two year of after-data excluded. large extent by a significant increase in
Measure. Stephen H. Richards, were collected and analyzed. Little conclusive data accident frequency and severity.
(1981), TRR 811, pp. 23–25. on public reaction to
Austin, Texas. light cut back.
RP-8 Assumption/
9 13 Fifteen Years of Road Data were studied during the 12 months In all cases saving measures unfavorably
Lighting in Belgium. prior to and 12 months following the influenced the accident patterns.
G. De Clercq, (1985), introduction of energy saving measures.
ILR 36(1), pp. 2–7.
10 14 Accident Analysis on Two The main findings of two accident Overall accident rates increase during
Non-Controlled Access analysis studies conducted on two winter, week ends, at night and at the
National Highways in Greece. highway sections of Greek National period before sunset. It is considered more
John M. Frantzeskakis, (1983), Roadway Network are presented. appropriate to correlate accident rates to
ITE Journal 53(2). pp. 26–32. levels of service instead of traffic
volumes.
11 18 Subjective Ratings of A series of panel rating experiments were The roadway surface VI and VL are highly correlated with
Visibility and Alternative conducted that provided measurements of was new bituminous MPR but for VI, both target size and
Measures of Roadway the subjective appraisal of visibility for a concrete. The (log) contrast conditions, and for VL, only the
Lighting. Michael S. Janoff, wide range of lighting conditions. transformed linear target contrast affects the form of
(1988), Journal of the IES regression technique relationship.
18 (Winter 1989), pp. 16–28, has been applied in all Contrast has a much larger effect on
Christianburg, Virginia. analyses relating to VI visibility than luminance.
and VL to mean panel
rating (MPR) which
may not be the best
form.
12 19 The Effect of Visibility A dynamic experiment was conducted to All luminance Contrast, VI and VL are highly correlated
on Driver Performance: determine two driving performance measurements were with DD and RD.
A Dynamic Experiment. measures; detection distance (DD) and made at fixed distances It appears that VL=2(VI) for small targets.
Michael S. Janoff, (1989), recognition distance (RD). from the target. Relationships are better if the data sample
Journal of the IES 19 (Winter is restricted to only negative contrast
1990), pp. 57–63. conditions.
Philadelphia, Pennsylvania. VL is superior under 2-200 watt HPS per
pole 200 ft opposite system.
13 The effect of Headlights The relative effect of roadway lighting All measurements At a distance greater than 275 ft there is
on Small Target Visibility. and roadway lighting in combination with were made at distances generally no effect of headlights. At
Michael S. Janoff, (1992), IES low beam vehicular lighting on from the target of 75 to distance less than 275-ft headlights may
Journal 21(2), pp. 46–53 photometric visibility and on distance at 775 ft. increase recognition distances in positive
Philadelphia, Pennsylvania. which drivers can detect and determine contrast but no such relationship found for
the orientation of a small target placed on road negative contrast.
A-21
Appendix A
surface were determined.

RP-8 Assumption/
A-22
14 Standard Target Contrast: A less sophisticated parameter to 20 cm × 20 cm middle With symmetrical luminaries the over
A Visibility Parameter Beyond represent a visibility index that can easily gray card (perfect lapping of the main beams is important to
Luminance to Evaluate the be computed at each road location on the diffuser) of 20 percent avoid spots with low or zero contrast.
Quality of Roadway Lighting. roadway surface by using only the reflectance is kept Increased negative contrast can be
F. W. Jung and A. Titishov, physical dimensions and properties of the 80–100 m away from achieved with non-symmetrical luminaires
(1987), TRR 1111, pp. 62–71. lighting system, is proposed. the driver. directed toward the driver without having
Ontario Ministry of Transp. & main beams touch or overlap and with
Communications, Canada. good distribution of luminance.
15 A Proposed Procedure for A method was developed to specify To measure VLeff, Disability glare factor to measure VLeff
Predicting Performance roadway lighting required to provide the transient adaptation has little effect.
Aspects of Roadway Lighting visibility needed for driver performance factor (for moving eye) Physical contrast measures based on
in Terms of Visibility. O. M. by using a 15:1 scale model roadway in is neglected because of “average contrast” formula (excluding
Blackwell and H. R. Blackwell, both concrete and bituminous pavement. the absence of an shadow) is found to be the best predictor
(1976), Journal of the IES agreed upon method to of visibility.
(April 1977), pp. 148–166. measure it.
Institute for Research in Vision,
Ohio State Univ., Columbus, Ohio.
16 Reflection Properties of Road This investigation concerns the The surface can be SL will be the same function of the angles
Surfaces in Headlight dependence of the geometry on the divided into a large for most road surfaces. SLs for humid
Illumination. Dependence specific luminance of road surfaces in number of small, plane conditions are a fraction of the SLs for
on Measuring Geometry. headlight illumination. The analysis of facets. the dry condition. The accuracy of the
National Swedish Road and data is based on model considerations of complete description system, as based on
Traffic Institute, (1982), Nordic the reflection leading to expressions for an assumed standard geometry of 50-m
Research Cooperation for the influence of geometry. spacing is found to be acceptable.
Night Traffic, Report No. 4,
1982, Linkoping, Sweden.
17 Methodology for Determining This paper describes the instrumentation No presentation of data Study incorporated factors such as surface
Pavement Reflectivity for and methodology for making automated or analysis results macrotexture and microtexture, aggregate
Roadway Luminance in-situ pavement reflectance because problems with size, surface finish, grading, mixture
Calculation, R. N. Helms measurements using gonio-reflectometer. reflectometer electronics design and climatic conditions.
(1983), TRR 904, invalidated the data. Recommends reporting of complete
Univ. of Colorado, Boulder, hemispherical data rather than reduced
Colorado. luminance coefficients.
RP-8 Assumption/
18 Light Reflection by Road This paper is the result of a long-term A new pavement reflectance parameter
Surfaces. Mededeling 53, study of the relationship between the light system SCW-KEMA, (similar to the
(1984), Germany. reflection properties and civil engineering CIE-R and CIE-N systems) is introduced.
properties of the pavement. It is recommended that a one- or two-
423 dry pavement surfaces in Eindhoven, parameter system is sufficient to
Copenhagen and Berlin were investigated. characterize a pavement. These parameters
can be S1 and QP.
The authors also recommend that two
groups of road surfaces (instead of four in
CIE-R) be used.
From a photometric standpoint, a preferred
road surfaces can be defined by S1 0.4
and QP ≤0.8 Cd/m2/lux.
These properties are given by:
• Asphalt pavements with at least 30
percent mineral aggregate consisting
white chippings.
• Portland cement concrete pavements
• Surface treatments with white chippings.
19 A Standard Practice for Alternative approaches to Table 2 Assumption: Equations were developed for negative
Roadway Lighting Based on luminance values in IES RP-8 (1983) are A maximum of 80 contrast relating required visibility index
Driver Performance. Michael considered because existing values are not percent performance with driver performance level (reliability).
S. Janoff, Journal of IES, 19(1), based on proven research. The alternative (reliability) is required
1990. approaches are based on panel rating for complex driving
experiments conducted in Christianberg, conditions.
Virginia, in 1987. Also, field experiments
were conducted in Philadelphia to
determine the time separation gap or TTT.
20 Visibility Response Expressions for total stopping distance is A one-second Equations are developed for a two-second
Distance. Michael S. Janoff, developed based on speed, VI and VL. correction to the reaction and response time in addition to
Journal of IES, 22(1), 1993. reaction time is the braking time (distance).
adopted to convert
Gallagher’s data
(drivers not alerted) to
compare with Janoff’s
A-23
Appendix A
data (drivers alerted).

A.5 BIBLIOGRAPHY
This bibliography is arranged chronologically.

1. Wilken, D., Burkett, K. et al. (September 2001). European Road Lighting Technologies. U.S.
Department of Transportation. Federal Highway Administration, International Technology
Exchange Program. FHWA-PL-01-034.
2. Green, Bobby, Olkan Culvaci, Sanjaya Senadheera, Karl Burkett, and Gransberg Douglas (October
2002). Evaluation of Roadway Lighting Systems Designed by Small Target Visibility (STV)
Methods. Final Report No. 0-1704-8, Department of Civil Engineering Texas Tech University
Lubbock, Texas 79409.
3. Illuminating Engineering Society of North America. (2000), American National Standard Practice
for Roadway Lighting, ANSI/IES RP-8-2000, Illuminating Engineers Society of North America,
New York, New York.
4. Ministry of Transport, Public Works and Water Management, Directorate-General of Public Works
and Water Management, Dynamic Public Lighting. (March 1999). Transport Research Centre
(AVV), P.O. Box 1031, 3000 BA Rotterdam, The Netherlands.
5. Adrian, W. K. et al. (1997) “Amendments in Calculating STV: Influence of Light Reflected from
the Road Surface on the Target Luminance.” Unpublished Data.
6. Adrian, W. K. (1997). “A Method to Predict Visual Functions and Visual Performance in Mesopic
Lighting Levels, “ Report prepared for the EPRI Lighting Research Office.
7. Adrian, W. K., (1997). “A Method to Predict Visual Performance in Mesoptic Lighting.” Report for
EPRI Lighting Research Office, Electric Power Research Institute, Palo Alto, California.
8. Adrian, W. K., and Gibbons, Ronald B. (1996). “Influence of Observation Angle on Road Surface
Reflection Characteristics.” Proceedings Paper #70, 1996 Annual Conference Illuminating
Engineers Society of North America, New York, New York.
9. Ellis, R. D., and Herbsman, Z. (1996). “Illumination Guidelines for Nighttime Highway Work.”
NCHRP Research Results Digest, No. 216, Transportation Research Board, Washington, D.C.
10. Lewin, I. (1996). “Advances in Measurement Technology for Vehicle Lighting Systems,” Society of
Automotive Engineering, Inc., pp. 107–114.
11. Lewin, I., Lin, F., and Sisson, C. (1996). “Accuracy Analysis of Video-Based Photometry,” Society
of Automotive Engineering, Inc., Society of Automotive Engineering, pp. 1–10.
12. Van Dommelen, C. (1996). “Choosing the Right Lighting for Inspection.” Test and Measurement
World, p. 53.
13. Swanlen, H., and Schnell, T. (1995). “Loss of Visibility Distance Caused by Automobile
Windshields at Night.” Transportation Research Record 1495, Transportation Research Board,
Washington, D.C., p. 128.
14. Texas Department of Transportation, (1995). Highway Illumination, Traffic Operations Manual,
Texas Department of Transportation, Austin, Texas.
15. Kahl, K. B., and D. B. Fambro, (1994). “Investigation of Object-Related Accidents Affecting
Stopping Sight Distances,” Transportation Research Record 1500, Transportation Research Board,
Washington, D.C., pp. 25–30.
16. Kasturi, R. (1994). “A Model-Based Approach for Detection of Runways and Other Objects in
Image Sequences Acquired Using an On-board Camera.” NASA Langley Research Center Report,
National Aeronautics and Space Administration, Langley, Virginia.
17. Persaud, B. N. and Mucsi, K. (1994). “Microscopic Accident Potential Models for Two-Lane Rural
Roads.” Transportation Research Record 1485, Transportation Research Board, Washington, D.C.,
pp. 134–139.
18. Zwahlen, H. T., and Schnell, T. (1994). “Loss of Visibility Distance Caused by Automobile
Windshields at Night.” Transportation Research Record 1495, Transportation Research Board,
A-24
Appendix A
19. Campbell, K. (1993). “Recent Research in Developing Accident Typologies.” Proceedings of

Human Factors Research in Highway Safety,
20. Freedman, M. (1993). “Effects of Reduced Transmittance Film on Automobile Rear Window
Visibility.” Human Factors, 35(3), pp. 535–550.
21. Janoff, M. S., (1993). “Toward Development of a Visibility Model for Roadway Lighting Design.”
Journal of the Illuminating Engineering Society, Illuminating Engineers Society of North America,
New York, New York, pp. 122–130.
22. Janoff, M. S., (1993). “Visibility vs. Response Distance.” Journal of the Illuminating Engineering
Society, 22(1), Illuminating Engineers Society of North America, New York, New York.
23. Janoff, M. S. (1993). “Measured versus Computer-Predicted Luminance in Roadway Lighting
Applications.” Journal of the Illuminating Engineering Society, Illuminating Engineers Society of
North America, New York, New York, pp. 75–82.
24. Keck, M. E., (1993). “Optimization of Lighting Parameters for Maximum Object Visibility and Its
Economic Implications.” Proceedings, 2nd International Symposium on Visibility and Luminance
in Roadway Lighting, Lighting Research Institute, New York, New York.
25. Zwahlen, H. T. (1993). “Effects of Automobile Windshields or Other Optical Filters on the
Visibility and Legibility of Targets Within the Driving Context.” Transportation Research Record
94, Transportation Research Board, Washington, D.C.
26. Elvik, R. (1992). “Meta-Analysis of Evaluations of Public Lighting as Accident Countermeasure.”
Transportation Research Record 1485, Transportation Research Board, Washington, D.C., pp. 112–
123.
27. Campbell, K. (1991). “Accident Typologies for Collision Reduction.” VTI Report 372A, Part I,
Linkoping.
28. Hostetter, R. S. et al. (1991). “Trade-Off Between Delineation and Lighting on Freeway
Interchanges: Investigation on Transient Visual Adaptation.” Report No. FHWA-RD-91-041, U.S.
Department of Transportation, Federal Highway Administration, Washington, D.C.
29. Janoff, M. S. (1991). “The Effect of Headlights on Small Target Visibility.” Journal of the
Illuminating Engineering Society, 21(2), Illuminating Engineers Society of North America, New
York, New York, pp. 46–53.
30. Adrian, W. K. and R.V. Topalova (1990). “Visibility Under Transient Adaptation.” Transportation
Research Record 1327, Transportation Research Board, Washington, D.C., pp. 47–48.
31. American Association of State Transportation and Highway Officials, (1990). A Policy on
Geometric Design of Highways and Streets, AASHTO, Washington, D.C.
32. Janoff, M. S., (1990). “A Standard Practice for Roadway Lighting Based on Driver Performance.”
Journal of the Illuminating Engineering Society, 19(1), Illuminating Engineers Society of North
America, New York, New York.
33. Janoff, M. S., (1990). “The Effect of Visibility on Driver Performance: A Dynamic Experiment.”
Journal of the Illuminating Engineering Society, Vol. 9, Illuminating Engineers Society of North
America, New York, New York, pp. 57–63.
34. Hair, E. A. and B. L. Green, (1990). “Component Efficiencies from the Operation of the Crosbyton
Solar Bowl.” Proceedings, 25th Intersociety Energy Conversion Engineering Conference, Dallas,
Texas.
35. Zwahlen, H. T., and J. Yu (1990). “Color and Shape Recognition of Reflectorized Targets Under
Automobile Low-Beam Illumination at Night.” Transportation Research Record 1327, Transporta-
tion Research Board, Washington, D.C., pp. 1–7.
36. Adrian, W. (1989). “Visibility of Targets:Model for Calculation.” Lighting Research and
Technology, Vol. 22 (NO4), Great Britain.
37. Bodmann, H. W., and H. J. Schmidt (1989). “Road Surface Reflection and Road Lighting: Field
Investigations,” Lighting Research and Technology, 20(4), pp. 159–170.
38. Box, P. C. (1989). “Major Route Accident Reduction by Improvements,” ITE 1989 Compendium of
Technical Papers, pp. 307–310.
A-25
39. Janoff, M. S. (1989). “Subjective Ratings of Visibility and Alternative Measures of Roadway
Lighting.” Journal of the Illuminating Engineering Society, Vol 18, Illuminating Engineers Society
of North America, New York, New York, pp. 16–28.
40. Keck, M. E., (1989). “Effect of Luminaire Arrangement on Object Visibility.” Transportation
Research Board 1247, Transportation Research Board, Washington, D.C., pp. 17–22.
41. Strickland, S. G., and V. J. Nowakowski (1989). “The Older Driver: A Growing Concern in
Roadway Design and Operations.” ITE 1989 Compendium of Technical Papers, pp. 302–306.
42. Alferdinck, J. W. and P. Padomos (1988). “Car Headlamps: Influence of Dirt, Age, and Poor Aim
on Glare and Illumination Intensities.” Lighting Research and Technology, 20(4), pp. 195–198.
43. Basset, M. G., et al. (1988). “Measurement of Reflection Properties of Wet Pavement Samples.”
44. Hostetter, R. S. et al. (1988). “Trade-Off Between Delineation and Lighting on Freeway
Interchanges: Investigation on Transient Visual Adaptation, Report No. FHWA-RD-88-223, U.S.
Department of Transportation, Federal Highway Administration, Washington, D.C.
45. Green, B. L., O’Hair, E.A., Simpson, T., Krile, T., Jones, W. B., and W. P. Vann, (1987). “Solar
Bowl Subsystems Tests and Analyses,” DOE Technical Report, U.S. Department of Energy
Contract DE-AC04-83AL21557.
46. Jung, F. W., and A. Titishov (1987). “Standard Target Contrast: A Visibility Parameter Beyond
Luminance to Evaluate the Quality of Roadway Lighting.” Transportation Research Record 1111,
Transportation Research Board, Washington, D.C., pp. 62–71.
47. Freedman, M., Decina, L. E., Nick, J. B., and E. F. Farber (1986). “Reflective Characteristics of
Roadway Pavements During Wet Weather.” Final Report, Report No. FHA/RD-86/104, Federal
Highway Administration, Washington, D.C.
48. Griffin, R. and D. B. Woodham (1986). “Reduced Freeway Lighting. ”Report CDOH-DTP-R-86-
19, Colorado Department of Highways, Boulder, Colorado.
49. Janoff, M. S., Staplin, L. K., and J. B. Arens (1986). “The Potential for Reduced Lighting on
Roadways.” Public Road, 50(2 ), pp. 33–42.
50. Stein, B., Reynolds, J. S., and W. J. McGuiness (1986). Mechanical and Electrical Equipment for
Buildings, 7th Edition, John Wiley and Son, New York, New York, pp. 857–1114.
51. Basset, M. G., and F. W. Jung (1985). “Pavement Reflection Studies.” Transportation Research
Circular N297, Transportation Research Board, Washington, D.C., pp. 12–13.
52. Clear, Robert, and Sam Berman (1985). “Visibility, and Roadway Performance, 1985.” Journal of
the the Illuminating Engineering Society, 5(2), Illuminating Engineers Society of North America,
53. De Clercq, G., (1985). “Fifteen Years of Road Lighting in Belgium.” International Lighting Review
36(1), pp. 2–7.
54. Freedman, M., Decina, L. E., and K. Y. Knoebel (1985). “Traffic Signal Brightness: An
Examination of Nighttime Dimming.” Report No. FHWA/RD-85/005, Federal Highway
Administration, Washington, D.C.
55. Ketron, J. B., (1985). “Reflective Characteristics of Roadway Pavements During Wet Weather.”
Transportation Research Circular N297 (Dec 1985), Transportation Research Board, Washington,
D.C., pp. 11–12.
56. Lamm, R., J. H. Kloeckner, and E. M. Choueiri (1985). “Freeway Lighting and Traffic Safety—
A Long-Term Investigation.” Transportation Research Record 1027, Transportation Research
Board, Washington, D.C., pp. 57–63.
57. Loch, H. C. (1985). “Roadway Lighting Calculations in Perspective.” Journal of the Illuminating
Engineering Society, Vol. 12, Illuminating Engineers Society of North America, New York, New
York, pp. 663–669.
58. Sanderson, J. T. (1985). “An Analysis of Accidents on Freeways With and Without Lighting.”
Report No. NTS85/2, Royal Automobile Club of Victoria, Australia.
A-26
Appendix A
59. Shelby, B. L. and B. G. Howell (1985). “Comparative Analysis of Roadway Lighting Using Both
CIE 19/2 and Adrian’s Delta L.” Journal of the Illuminating Engineering Society, Illuminating
Engineers Society of North America, New York, New York, pp. 181–211.
60. Anderson, Kyle A., W. J. Hoppe, P. T. McCoy, and R. E. Price (1984). “Cost- Effectiveness
Evaluation of Rural Intersection Levels of Illumination.” Transportation Research Record N995,
61. Jung, W., Kazakov, A., and A. L. Titshov (1984). “Road Surface Reflectance Measurements in
Ontario.” Transportation Research Record 996, Transportation Research Board, Washington, D.C.,
pp. 24–37.
62. Basset, M. G., et al. (1983). “Automated Facility for Measurement of Pavement Sample Reflectance
Characteristics,” Transportation Research Record 904, Transportation Research Board,
63. Frantzeskakis, J. M. (1983). “Accident Analysis on Two Non-Controlled Access National
Highways in Greece.” Institute of Traffic Engineers Journal, 53(2), pp. 26–32.
64. Helms, R. N. (1983). “Methodology for Determining Pavement Reflectivity for Roadway
Luminance Calculation.” Transportation Research Record N904, Transportation Research Board,
65. Illuminating Engineering Society of North America, (1983). “Proposed American National
Standard Practice for Roadway Lighting.” Journal of the Illuminating Engineering Society, 12(2),
Illuminating Engineers Society of North America, New York, New York, pp. 146–196.
66. Janoff, M. S., Freedman, M., and L. E. Decina (1983). “Partial Lighting of Interchanges,” NCHRP
Report 256, National Cooperative Highway Research Program, National Research Council,
Washington, D.C.
67. National Cooperative Highway Research Program (1983). “Practical Guide for Minimizing Tort
Liability.” NCHRP Synthesis 106, Highway Research Board National Research Council,
Washington, D.C.
68. Olson, P. L., and M. Sivak (1983). “Improved Low-Beam Photometrics.” National Highway Traffic
Safety Administration, Report No. DOT HS-9-02304, U.S. Department of Transportation, National
Highway Traffic Safety Administration, Washington, D.C.
69. Olson, P. L., and M. Sivak (1983). “Improved Low-Beam Photometrics.” 1983, Report No. DOT
HS-806-487, Department of Transportation, National Highway Traffic Safety Administration,
Washington, D.C.
70. Finch, D. M. (1982). “Roadway Visibility Using Minimum Energy.” Transportation Research
Record 855, Transportation Research Board, Washington, D.C., pp. 7–12.
71. Freedman M., and P. S. Davit (1982). “Improved Conspicuity to the Sides and Rear of Motorcycles
and Mopeds.” Final Report, Report No. DOT HS-806-377, Department of Transportation, National
Highway Traffic Safety Administration, Washington, D.C.
72. CIE (1981). “An Analytic Model for Describing the Influence of Lighting Parameters Upon Visual
Performance.” CIE Report No. 19/2, International Commission on Illumination.
73. Richards, S. H. (1981). “Effects of Turning Off Selected Roadway Lighting as an Energy
Conservation Measure.” Transportation Research Board Record N811, Transportation Research
74. Freedman, M. (1980). “National Accident Sampling System: Establishment and Operation of the
Primary Sampling Unit in Delaware County, PA, 1977–80.” Report No. DOT-HS-805 275,
Department of Transportation, National Highway Traffic Safety Administration, Washington, D.C.
75. Los Angeles Bureau of Street Lighting: (1980). “Intersection Traffic Accident Reduction Through
Improved Street Lighting.” OTS Project 127803 Report, Department of Public Works, City of Los
Angeles, California.
76. Scott, P. P. (1980). “The Relationship Between Road Lighting Quality and Accident Frequency.”
TRRL Laboratory Report 929. Department of Transportation, National Highway Traffic Safety
Administration, Washington, D.C.
A-27
77. U.S. Environmental Protection Agency, (1980). Interim Guidance for Visibility Monitoring,
Environmental Monitoring System Laboratory, U.S. Environmental Protection Agency,
Washington, D.C.
78. Merritt, J. O., Newton, R. E., Sanderson, G. A., and M. L. Seltzer (1979). “Driver Visibility Quality:
An Electro-Optical for In-Vehicle Measurement of Module Transfer (MTF).” U.S. Department of
Transportation, National Highway Traffic Safety Administration, Washington, D.C.
79. Nielsen, B. et al. (1979). “Effects of Wear and Composition on Road Surface Reflection Properties.”
CIE Proceedings (19th session), International Commission on Illumination, pp. 425–432.
80. Scott, P. P., Cobb, J., Hargroved, R. A., and A. M. Marsden (1979). “Road Lighting and Accidents.”
CIE Proceedings, International Commission on Illumination, Kyoto, Japan.
81. Hall, R. R., and A. J. Fisher (1978). “Measures of Visibility and Visual Performance in Road
Lighting.” ARRB Research Report No. 74, Australian Road Research Board, Victoria, Australia.
82. Hilton, M. H. (1978). “Continuous Freeway Illumination and Accidents on a Section of Route I-95.”
Report VHTRC 79-R4, Virginia Highway & Transportation Research Council, Richmond, Virginia.
83. Koth, B. et al. (1978). “Vehicle Fog Lighting: An Analytical Evaluation.” Report No. DOT-HS-803
442, Department of Transportation, National Highway Traffic Safety Administration, Washington,
D.C.
84. National Cooperative Highway Research Program (1978). “Cost and Safety Effectiveness of
Highway Design Elements.” NCHRP Report No. 197. Highway Research Board National Research
Council, Washington, D.C.
85. Van Bommel, Wout J.M., (1978). “Design Consideration for Roadway Lighting.” Journal of the
Illuminating Engineering Society, Illuminating Engineers Society of North America, New York,
New York, pp. 40–46.
86. Blackwell, O. M., and H. R. Blackwell (1977). “A Proposed Procedure for Predicting Performance
Aspects of Roadway Lighting in Terms of Visibility.” Journal of the Illuminating Engineering
Society, Illuminating Engineers Society of North America, New York, New York, pp. 148–166.
87. Fisher, A. J. (1977). “Road Lighting as an Accident Counter-Measure.” Australian Road Research,
7(4), Australian Road Research Board, Victoria, Australia, pp. 2–16.
88. Gordon, P. (1977). “Appraisal of Visibility on Lighted Dry and Wet Roads.” Lighting Research and
Technology 9(4), pp. 177–188.
89. Janoff, M. S., Koth, B., McCunney, Freedman, M., Duerk, and M. Berkovitz (1977). “Effectiveness
of Highway Arterial Lighting.” 1977, FHWA-RD-77-37, Federal Highway Administration,
Washington, D.C.
90. Ketvirtis, A. and P. J. Cooper (1977). “Detection of Critical Size Object as a Criterion for
Determining Drivers Visual Needs.” Proceedings, 76th Annual Meeting of the Transportation
Research Board, Washington, D.C.
91. Moerman, J.J.B. (1977). “Accuracy of Photometry of Retroreflectors and Retroreflective
Materials.” Journal of the Illuminating Engineering Society, 9(2), Illuminating Engineers Society
of North America, New York, New York, pp. 85–91.
92. CIE (1976). “Calculation and Measurement of Luminance and Illuminance in Road Lighting.” CIE
Publication 30, International Commission on Illumination.
93. Gallagher, Vincent P. (1976). “A Visibility Metric for Safe Lighting of City Streets.” Journal of the
Illuminating Engineering Society, 5(2), Illuminating Engineers Society of North America, New
York, New York, pp. 85–91.
94. Ellis, K. L. (1976). “Measurement of Directional Reflectance of Pavement Surfaces and
Development of Computer Techniques for Calculating Luminance.” Journal of the Illuminating
Engineering Society, 5(2), Illuminating Engineers Society of North America, New York, New York,
pp. 118–126.
95. Frederiksen, E. and K. Sorensen (1976). “Reflection Classification of Dry and Wet Road Surfaces.”
Lighting Research and Technology, 8(4 ), pp. 175–186.
A-28
Appendix A
96. Holmes, J. G. (1976). “The Road Surface Is Part of the Road Lighting.” Light, Lighting and
Environmental Design, London, England, p. 204.
97. Lewis, Alan L. (1976). “The effects of Illumination and Contracts on Visual Processes Affecting
Comfort and Performance.” Lighting Design and Applications, 6(1), Illuminating Engineers
Society of North America, New York, New York, p. 16.
98. Marsden, A. M. (1976). “Road Lighting—Visibility and Accident Reduction.” Public Lighting, 41
(175), Association of Public Lighting Engineers, pp. 106–111.
99. Waldram, J. M. (1976). “Safety on the Road at Night.” Light and Lighting and Environmental
Design, London, England, pp. 184–187.
100. Wright, W. D., (1976). “Seeing to Drive at Night,” Light and Lighting and Environmental Design,
London, England pp. 188–189.
101. Gallagher, V., Koth B., and M. Freedman (1975). “The Specification of Street Lighting Needs.”
Report No. FHWA-RD-76-17. Federal Highway Administration, Washington, D.C.
102. Rackoff, N. J. and T. H. Rackwell (1975). “Driver Search and Scan Pattern in Night Driving.” Ohio
Department of Transportation and The Federal Highway Administration, Special Report 156.
Columbus, Ohio.
103. Walton, N. E. (1975). “Fixed Illumination as a Function of Driver Needs.” Ohio Department of
Transportation and The Federal Highway Administration, Special Report 156. Columbus, Ohio.
104. Rinalducci, E. J. et al. (1974). “Losses in Nighttime Visibility Caused by Transient Adaptation.”
105. Helmers, G. and K. Rumar (1973). “Obstacle Visibility in Rural Night Driving as Related to Road
Surface Reflective Qualities.” Transportation Research Record N502, Transportation Research
106. Box, P. C. (1972). “Comparison of Accidents and Illumination.” Highway Research Record 416,
107. King, E. L. (1972). “Luminance Versus Luminance.” Highway Research Board Special Report 134,
108. Meese, G. E. (1972). “Vehicular Lighting Systems for Two-Lane Rural Highways.” Highway
Research Board Special Report 134, Transportation Research Board, Washington, D.C., pp. 41–65.
109. Box, P. C. (1971). “Relationship Between Illumination and Freeway Accidents.” Journal of the
Illuminating Engineering Society 66(5), pp. 365–393.
110. Hochreither, F. C., (1969). “Analysis of Visibility Observation Methods.” System Development
Office Test and Evaluation Laboratory Report, U.S. Weather Bureau, Sterling, Virginia.
111. King, L. E. and D. M. Finch (1968). “A Laboratory Method for Obtaining Pavement Reflectance
Data.” Highway Research Record No. 216, Transportation Research Board, Washington, D.C.,
pp. 23–33.
112. Strickland, J., Ward, B., and M. J. Allen (1968). “The Effect of Low- vs. High-Beam Headlights and
Ametropia on Highway Visilbility at Night.” American Journal of Optometry and Archives of
American Academy of Optometry, 45(2).
113. Finch, D. M., King, L., and K. L. Ellis (1967). “A Simplified Method for Obtaining Pavement
Reflectance Data.” Highway Research Record 179, Transportation Research Board, Washington,
D.C., pp. 53–60.
114. National Cooperative Highway Research Program (1967). “Effects of Illumination on Operating
Characteristics of Freeways.” NCHRP Report No. 60. Highway Research Board, National Research
Council, Washington, D.C.
115. Richards, O. W. (1967). “Visual Needs and Possibilities for Night Driving: Part 2.” The Optician,
154(3999), p. 523.
116. Richards, O. H. (1966). “Vision at Levels of Night Road Illumination.” Transportation Research
Record 179, Transportation Research Board, Washington, D.C., pp. 61–66.
A-29
117. Waldram, J. M. (1966). “The Design of the Visual Field in Street: The Visual Engineer’s
Contribution.” Transactions, 31(1 ), Illuminating Engineers Society of North America, London,
England.
118. Fitzpatrick, J. T. (1960). “Unified Reflective Sign, Pavement, and Delineation Treatments for Night
Traffic Guidance.” Highway Research Board Bulletin No. 255, Transportation Research Board,
119. Haber, H. (1955). Safety Hazard of Tinted Automobile Windshields at Night.” Optical Society of
America, 12(6 ), pp. 1–15.
120. Christie, A.W. (1954). “Reflection Characteristics of Pavement Surfaces.” Highway Research
Board Bulletin No. 89, Transportation Research Board, Washington, D.C., pp. 21–37.
121. Heath, W. and D. M. Finch (1953). “The Effect of Tinted Windshields and Vehicle Head Lighting
on Night Visibility.” Bulletin 68, Highway Research Board, National Research Council,
Washington, D.C.
122. Roper, V. J. (1953). Bulletin 68, Highway Research Board, National Research Council, Washington,
D.C.
123. Finch, D. M. and A. E. Simmons (1950). “Uniformity of Illumination in Highway Lighting.”
Illuminating Engineering 45, p. 561.
124. Blackwell, H. R. (1946). “Contrast Thresholds of the Human Eye.” Journal of the Optical Society
of America, 36(11).
125. Waldram, J. M. (1938). “The Revealing Power of Street Lighting Installation.” Transactions,
Illuminating Engineers Society of North America, London, England.
A-30
Appendix B
Lighting Basics
Purpose of Roadway Lighting
The general purpose of roadway lighting is to provide improved safety, security, and aesthetics for the
various users of the roadways and associated facilities.
User. The term users may include vehicle operators (autos, trucks, buses, motorcycles, bicycles),
pedestrians, and other citizens of the community.
Roadway. Roadways are defined as freeways, highways, and city streets of various types.
Associated Facilities. Associated facilities include appurtenances of a traffic and non-traffic nature.
Traffic appurtenances include physical features along the roadway such as traffic barriers, bridge piers,
roadside ditches, curbs, channelization, etc.
Identifying roadways that need lighting and designing lighting systems to improve driver visibility
requires an understanding of what drivers need to see. Driving safely and efficiently requires a constant
flow of information relative to the roadway, traffic, and environmental conditions.
Physical Characteristics of Light and Vision
Light as Radiant Energy
Light may be defined as that portion of the electro-magnetic spectrum detectable by human vision—
between 380 nanometers (nm) and 780 nm. The detection of radiant energy within those limits depends
on many factors, including the relative sensitivity of the eye to the incident flux. While the unit of radiant
power is the watt, the unit used to describe the effective power of light is known as the lumen (lm). The
lumen is a measure of the time rate of flow of radiant energy into the eye, multiplied by the sensitivity of
the eye, for each wavelength of the incoming radiant energy.
Knowledge of the luminous flux (total lumen output) of a source does not provide information on
the distribution and eventual usefulness of the emitted light, nor is it instructive regarding our perception
of the light source. The luminous intensity of a light source is a measurement of the luminous flux of the
source in a given direction, per unit solid angle (steradian), in units of candela (cd). The illuminance is a
measurement of the total luminous flux striking or passing through a given area. Illuminance is measured
in lumens per area—either lux (lm/m2) or footcandles (lm/ft2), abbreviated (fc). Thus, the amount of light
striking a road surface from a system of fixed roadway lighting luminaires is the illuminance provided by
the luminaires. When light is emitted from an extended area, such as from a fluorescent lamp fixture or
light reflected off a wall or road surface, the eye cannot sum all of the observed flux into a single value of
luminous intensity. Instead, the eye reacts to the luminous intensity emitted in the direction of the
observer over some discrete observation area. For measurement purposes, we speak of the luminance of
a source, which is the luminous intensity per unit area of the source, in units of cd/m2 or cd/ft2. Luminance
is often described as the “brightness” of an object; however, the two are very different. Luminance is a
B-1
physical phenomenon, and is independent of the observer or surrounding conditions. Brightness is a

physiological phenomenon used to describe the sensation of what is perceived. It depends on the adapta-
tion of the eye and the relative luminance of the object and its surroundings. For example, the moon looks
bright against the dark night sky and dim when seen against the daytime sky but its measured luminance
is a constant value. When assessing colors, the saturation of the color—that is, how nearly monochro-
matic the stimulus is—also influences the sensation of brightness.
Note that luminous intensity is a measurement of the light emitted from a source in a given direction,
per steradian, and does not change with the distance from the source. The illuminance provided by a light
source is, however, dependent on the area illuminated. As the source is moved towards or away from a
given surface the illuminated area changes and the illuminance, likewise changes. The luminance of a
source is the luminous intensity emitted by a unit area of the source in a given direction, and does not
change with distance.
Transmittance
The luminous transmittance of a substance is a measure of its capability to transmit light through the
material. The nomenclatures for luminance transmittance are:
• transmittance
• transmission factor
• coefficient of transmission
• transmission coefficient.
These terms are used interchangeably. The transmittance is the ratio of the total transmitted light to the
total incident light. Transmittance must be used cautiously because materials may be wavelength selec-
tive in transmitting light; so a spectral analysis of incident and transmitted light is sometimes called for if
a material is selective in a wavelength of interest. For this study, the wavelengths of interest are restricted
to visible light, so we will easily recognize a wavelength selective filter. In general the transmission coef-
ficient should refer to materials displaying non-selective absorption characteristics.
Reflectance
The ratio of reflected light to incident light is referred to by any of the following:
• reflectance
• reflectance factor
• reflectance coefficient.
Reflectance is a measure of the light that bounces off a surface and is not transmitted. If half of the inci-
dent light is bounced off the surface, the surface reflectance coefficient is 0.5 or 50 percent. If a beam of
light reflects from a smooth surface, the reflection is known as specular and reflects away from the sur-
face as a single beam of light. If the surface is very rough, the reflections for a beam of light are scattered
by the multifaceted surface. The light reflects in all directions away from the surface, and the reflection
is called diffuse.
Refraction
The speed of light in a material, vm, and the speed of light in free space, co, are related to the index of
refraction, n, by the following:
co
n⫽ (B-1)
vm
B-2
Appendix B
The index of refraction, n, is always greater than 1; therefore, vm is always less than, co.
Figure B-1. Specular ray tracing model.
• Incidence radiation, Io, @ Incidence angle, θi,

• Reflected radiation, Irfl, @ Reflected angle, θrfl,
• Transmitted radiation, It, @ Refracted angle, θrfa,
• The angle of incidence is equal to the angle of reflection, vi = θrfl.
Refraction takes place at a boundary where indices of refraction change. The incident angle and the
refracted angle are related by Snell’s Law, and reflect differences in speeds of light in the respective
mediums.
Snell’s Law: n1 sin θ1 = n2 sin θ2.
Transmission angle, θt,
Absorption
Each time refraction takes place at a boundary, a portion of the incident light passes from medium, n1,
to medium, n2, and the portion is transmitted. If the material is glossy some of the energy is converted
from visible radiation to infrared radiation (heat) and lost (from the visible spectrum). The losses are
absorption losses.
Absorption losses are exponential with distance such that
I(x) = Ioe2kx (B-2)
where:
Io is the incident radiation entering the material
x is the distance traveled through the material
k is the loss coefficient for the material
e is 2.718281 . . .
Absorption losses are losses due to energy transformation from higher energy, visible light to lower
energy, infrared non-visible light. Changing the radiation from the visible spectrum to the non-visible
spectrum is thought of as a loss to the visible spectrum and a loss to an observer.
Diffusion
Diffuse reflections are due to first surface roughness and reflections at the boundary surface.
B-3
Figure B-2. First surface diffuse reflections.
Diffuse transmissions are due to second surface roughness and refraction at the second surface boundary.
Figure B-3. Second surface diffuse refraction and transmissions.
Figure B-4. Lambertian reflection or refraction/transmission distribution.
Lambertian distribution, I(θ) = Imax cos(θ), is a diffuse reflection distribution or refraction distribution
due to the surface characteristics of a material. The roughness of the surface determines the reflection and
refraction directions.
Surfaces are not flat, so the reflections, refractions, and transmissions have a partial specular charac-
teristic and a partial diffuse characteristic as shown in Figure B-5.
Figure B-5. Reflection and transmission distributions.
B-4
Appendix B
Most surfaces are somewhat smooth and somewhat rough, so we get both a diffuse reflection and a
specular reflection. The reflectance is a measure of the total light reflected from the surface of any mate-
rial. The reflectance does not depend on whether the surface is diffuse or specular; all the reflected light
is measured. The ratio of incident light lost in a material is called the absorption coefficient. The absorbed
light is not lost, it is simply changed from visible wavelengths to lower energy, non-visible wavelengths
usually in the infrared. The sum of the transmitted, reflected, and absorbed light is equal to the incident
light. The transmitted light may also be diffused after it passes through some material, but the total
amount of light passing through the material is used in the transmission measurement to determine the
transmission coefficient. Just as the total reflected light is used in the reflectance measurement to obtain
the reflection coefficient.
Vision and Visibility
Humans rely primarily on visual information to interact with their environment.
Anatomy of the Eye
The process of vision begins with the eye, the sensory organ responsible for detection of light. The
photoreceptors within the eye are located in the retina, a thin covering on the back, inner surface of the
eye. There are two types of photoreceptors, rods and cones, which form the basis of two-receptor systems.
While the processes that link detection of a visual signal within the eye to comprehension of that sig-
nal by the brain are not well understood, there are models that provide good correlation between stimuli
and reported observations. One precept for these models is the duplicity theory of vision, which postu-
lates that the receptor systems of the rods and cones are independent, with their own particular properties.
Note that detection of a visual stimulus is evaluated over the entire receptor system, including the recep-
tors and the neural network that processes, encodes, and transmits signals to the brain.
Figure B-6. A horizontal cross-section of the human eye.
• Approximate length from cornea to fovea of retina is 24 mm.

• Thickness of choroid is about 0.05 mm, and the sclera 1.0 mm.
B-5
There are approximately 120 million rod receptor cells and 6.5 million cone receptor cells in the
human eye. The rod receptor system is more sensitive at very low light levels, but generally provides poor
visual acuity and no color discrimination. The peak sensitivity of the rod system is at a wavelength of
507 nm. The cone system is comprised of three different cone receptor cells (commonly called blue,
green, and red cones), with different sensitivity functions, and is responsible for color vision. The overall
sensitivity curve of the system is at about 555 nm. The cone system also provides relatively high visual
acuity, but requires significantly higher light levels to function.
Visual Functions of Rods and Cones
The three visual functions provided by the two receptor systems (rods and cones) are scotopic, pho-
topic, and mesopic vision. Descriptions of each follow.
Scotopic Vision. At luminance levels below about 0.05 cd/m2, there is insufficient light to activate the
cone receptor system, and vision is provided solely by the rod system. Scotopic vision is commonly
called “night vision.” There is no sensation of color in scotopic vision.
Photopic Vision. At luminance levels above 3 cd/m2, the rod system is effectively shut down through
interaction with the cone system, and cones are the principal active elements. Central (foveal) vision is
good, and normal color vision is possible.
Mesopic Vision. At luminance levels between those suited for scotopic and photopic vision, both recep-
tor systems provide visual information. Which system provides specific visual cues depends on the actual
luminance level and the relative spectral sensitivities of the two systems. Typically, at the lowest mesopic
levels, the cone receptor system only functions at longer wavelengths. As the luminance level increases,
the cone system essentially turns the “gain” down on the rod system, preventing glare and allowing the
cone system to predominate over increasingly shorter wavelengths. Thus, at very low mesopic levels, red
will be the first color a person can detect, with blue the last color detectable as the luminance level
increases.
Color Vision
A model of human color-vision is based on the opponents theory. This theory postulates that signals
from blue cones are compared against the summed signals of green and red cones to provide a value along
a blue-yellow axis. Signals from the green cones are compared against that of the red cones to provide a
value along a green-red axis. Finally, the summation of the green and red cone signals provides a lumi-
nance value. These three values allow the brain to identify the color in a color space. Note that although
there are three types of color receptors, there are only two chromatic channels and one luminance chan-
nel for our color vision system.
Distribution of Visual Sensitivity Within the Eye
Cone density is greatest in the fovea, a small area at the visual axis of the eye, and drops off rapidly to
a minimum at about 10 degrees of eccentricity. Cone density remains fairly uniform from 10 degrees
eccentricity to the limit of the retina. Rods are absent in the fovea, and then increase rapidly in number to
a maximum density at about 18 degrees of eccentricity. Beyond 18 degrees the rod density falls gradually,
but remains significantly higher than cone density. Clusters of rod and cone receptor cells are linked to
individual ganglion cells, which provide the actual signals to the brain. Thus, each “pixel” of the retina,
which creates an image of what we are observing, will have more than one receptor cell contributing to
B-6
Appendix B
the production of a neural signal. This clustering of receptor cells increases the sensitivity of the visual
system, but puts a limit on visual acuity. In the fovea, where there is a very high density of cone receptor
cells and no rods, the area occupied by a cluster of receptor cells is extremely small, resulting in the small-
est pixels within the retina. The fovea provides the highest resolution capability of the eye, and is the sec-
tion of the eye where we will instinctively try to focus a task-image. With no rods, and the smallest
number of receptor cells per ganglion cell, the fovea is also the least sensitive area of the retina, requiring
the highest level of luminance for vision and a complete insensitivity under scotopic conditions. Towards
the periphery of the retina, the clustered receptor cells occupy larger areas, resulting in lower image res-
olution.
Night Driving Conditions
Most night-driving situations occur under conditions of mesopic vision, although a well-designed
roadway lighting system can approach the photopic vision range in luminance.
Visibility
Visibility is the quality or state of being perceptible to the eye. In many applications, visibility is
defined in terms of the distance at which an object can be just perceived by the eye. In other situations,
visibility is defined in terms of the contrast or size of a standard test object, observed under standardized
viewing conditions, having the same threshold as the given object. Factors that directly influence visibil-
ity for drivers include:
• the luminance of objects on or near the roadway
• the luminance of the roadway and other backgrounds against which objects are viewed
• the size of an object and the identifying details of the object
• the contrast between an object, or the roadway, and is surroundings
• the time available for detection and recognition of objects
• the presence of glare from on- and off-roadway light sources (including other vehicular lighting)
• the driver’s physical and mental condition (including visual ability)
• the condition and cleanliness of a vehicle’s windshield.
TABLE B-1. Variables Associated with the Components of Visual Acuity
Component Primary Variables Secondary Variables
Task • size • type of object

• luminance • degree of accuracy required
• contrast • moving or stationary target
• exposure time • peripheral patterns
Lighting Conditions • illumination level • luminance ratios
• disability glare • brightness patterns
• discomfort glare • chromaticity
Observer • illumination level • subjective impressions
• disability glare • psychological reactions
• discomfort glare
B-7
Visual Acuity
There are three components to visual acuity in any seeing task: the task, the lighting conditions, and
the observer. Each of these acuity components has primary and secondary variables associated with it
(Stein et al., 1986), as shown in the following table.
Visual Size
The visual size of any detail that needs to be seen is a function of its physical size and its distance from
the point of observation. By combining these two dimensions, one can express the size as a visual angle,
which usually is measured in minutes of arc. Thus, the farther a given object is from the eyes, the smaller
its visual size becomes. Various types of test objects, three of which are shown in Figure B-7, have been
used for evaluating the size discrimination ability of the eye. In each case, the critical detail that must be
discriminated is indicated by the dimension d. So for a constant viewing distance, the visual angle sub-
tended at the eye by d is the same for the three objects, even though the maximum dimensions are different.
Another often-used way of expressing the size threshold of the eye is in terms of visual acuity.
Quantitatively, this is the reciprocal of the visual angle. It often is defined as the ability of the eyes to
resolve small detail. In driving, two types of visual acuity are of concern: static and dynamic visual acuity
(explanations follow).
Figure B-7. Commonly used test objects for determining size discretion and visual
acuity.
Static Visual Acuity. Static visual acuity occurs when both the driver and the object being viewed are
stationary. Static visual acuity is a function of background brightness, contrast, and time. With increasing
illumination, visual acuity increases up to a background luminance of about 32 cd/m2, and then remains
constant despite further increases in illumination. Static visual acuity also increases with increasing con-
trast of the object. Optimal exposure time for a static visual acuity task is 0.5 to 1.0 second when other
visual factors are held constant at some acceptable level.
Dynamic Visual Acuity. When there is relative motion between the observer and an object, such as
occurs in driving, the resolving ability of the eye is termed dynamic visual acuity. Dynamic visual acuity
is more difficult than static visual acuity because eye movements are not generally capable of holding a
steady image of the target on the retina. The image is blurred, and therefore, its contrast decreases. The
conditions favorable for dynamic visual acuity are slow movement, long tracking time, and good illumi-
nation. These are rarely found in the nighttime driving environment, but are very important for sign read-
ing, distance judgment, object discrimination, and other dynamic visual acuity tasks.
B-8
Appendix B
Contrast and Visibility Level
The contrast between an object and its surroundings is one of the key determinates of visual acuity.
The contrast (C) of an object to its background is defined as
Lo ⫺ Lb
C⫽ (B-3)
Lb
where Lo and Lb are the luminances of the object and the background, respectively. Recognition of an
object is most often based on a discrimination of the luminance differences. For nighttime conditions, an
obstacle may appear as a dark area against a bright background (negative contrast), or it may appear as a
–
bright area against a dark background (positive contrast). Contrast threshold, C/C , is the level of contrast
required for an object to be discernable, usually with a probability of detection of 50 percent, for a spec-
ified condition. The ratio of the actual contrast to the contrast threshold is designated the visibility level
(VL):
C
VL ⫽ (B-4)
C
A VL of less than one indicates that there is a low probability that an object will be detected. As VL
increases above a value of one, the probability of detection also increases. Note that even with a high
value of VL, there is no certainty of detection, rather only an increasing probability. Detection relies on
several factors other than contrast, including the amount of time available to make an observation, the
physical and mental state of the observer, and the presence of glare.
Illuminance and Luminance Evaluations
The calculation of the illuminance at a point, whether on a horizontal, vertical, or inclined plane, con-
sists of two parts: the direct component and the reflected component. The total of these two components
is the illuminance at the point in question. Of the methods for determining the direct component of the
illuminance, two methods—inverse square and illumination charts and tables—can be used for evaluat-
ing the effect of inclination.
Inverse-Square Method for Calculating Illuminance. Variations in the formula for the inverse-square
law are used to determine the illuminance at definite points, where the distance from the source is at least
five times the maximum dimension of the source. In such situations, the illuminance is proportional to the
square of the distance from the source.
The illuminance on a horizontal plane (Eh), as measured at a point defined by a set of angles from a
given light source (φ,γ), illustrated in Figure A-4, is expressed as:
I (f, g)cos g
Eh ⫽ (B-5)
D2
B-9
where:
I (φ,γ) = luminous intensity of the source in the direction of the point in question;
D = actual distance from the light source to the point in question;
H = vertical mounting height of the light source above the horizontal plane;
φ = the angle from the light source reference axis to the point in question; and
γ = the angle from the nadir of the light source to the point in question.
For a horizontal plane, cos γ = H/D. Therefore:
I (f, g)cos3 g
Eh ⫽ (B-6)
H2
Pavement Surface Luminance
The luminance of a surface (L) is defined as the luminous flux per steradians emitted (reflected) by a
unit area of surface in the direction of the observer. When the unit of flux per steradians is candela and the
area is measured in square meters, the unit of luminance is candela per square meter (cd/m2). The surface
luminance at a point can be calculated if the reflectance coefficient at the point, based on the incidence
angles, q(β,γ), and the illuminance value are known:
1
L⫽ Eh q(b, g) (B-7)
p
where:
Eh = illuminance at the point in question, as defined in Equation (B-6); and
q(β,γ) = the directional reflectance coefficient for angles of incidence β and γ.
By combining Equations (B-6) and (B-7), the luminance of the pavement surface can be calculated as:
q(b, g) I (f, g)cos3 g

L⫽ (B-8)
pH 2
Although a simple concept of the quantity of light reflected by a surface is assessed from the
reflectance coefficient, q(β,γ), the distribution pattern depends on the surface characteristics and the
angular relationship between the light source, the point in question, and the observer position. In princi-
ple, two types of reflectance are identified: diffuse and specular (mirror). New fallen, dry snow is an
example of a diffuse surface, whereas a smooth, wet road is a good example of a specular surface. Most
road surfaces provide a mixture of diffuse and specular reflectance.
In practice, q(β,γ)cos3 γ is expressed as a reduced coefficient of reflectance, R, and is given in a table
for each road surface classification (see IESNA RP-8-00, Tables A1-A4). Additional discussion of the
Coefficient of Pavement Reflectance, and the reduced coefficient of reflection, is found later in this
Appendix.
Accuracy of calculations of pavement luminance depends on the following factors:
• whether the photometric data used to determine the candlepower intensity at a particular angle
correctly represent the output of the lamp and luminaire
• whether the directional reflectance table accurately represents the reflectance of the actual surface
• atmospheric attenuation factors, such as mist or fog.
B-10
Appendix B
Field Evaluation of Installed Systems
The difference between predicted and calculated point luminance values can be substantial. One
study (Janoff, 1993) found “almost no consistent relationship between measured and predicted point-
wise luminances at 84 m (275 ft) distances.” And, “for luminances the differences between measured
and predicted can be attributed to a number of well known, but not easily quantifiable conditions. For
background luminances the differences result from the nonuniformity of the road surface—caused by
wear, dirt, and other surface contaminants; seasonal variations in moisture content; aggregate polish-
ing, etc.—and difference between the actual road surface and the CIE ideal road surface defined by an
R-table.”
Methods of Evaluation
The evaluation of pavement luminance and veiling luminance (explained later) of an installed system
can be done in three ways:
• direct measurement with luminance instruments
• measurement of incident light
• calculation from photometered luminaires.
Discussions of each of these methods follow.
Direct Measurement with Luminance Instruments. Luminance of the pavement can be measured with
a special telephotometer. The instrument should be set up at the observer’s position and a measurement
made at the desired point. If the moving observer method (IES) method of calculation has been used to pre-
dict the luminance at the same point, then a different instrument location is required for each point mea-
sured. In addition, it is very difficult to determine if the difference between calculated and measured values
is due to the lighting system or the pavement. While it is necessary to use this method in research investi-
gations it is not recommended as a method to determine if an installation is performing as specified.
Measurement of Incident Light. Measurement of incident light can be used as a method of determin-
ing if an installation is performing as specified. In this method a computer printout of the predictedinitial
incident light level (footcandles or lux) is made at the same time as the printout of the luminance. After
the installation is completed and operating, the incident light level can be measured at each grid point in
the customary manner. This method eliminates the pavement as a variable and will reveal incorrect instal-
lation procedures such as leveling, tilt, or rotation. These are difficult to separate from the effects of other
variables such as dirt, light source output, luminaire and ballast variability, and voltage variations.
Calculation from Photometered Luminaires. Calculation from photometered samples is an excellent

way to evaluate the performance of an installation. Several random samples of luminaires (with ballast)
and lamps are selected and sent to a competent laboratory to be photometered. The resultant photometric
data can be averaged (or analyzed independently) and the luminance calculated by a computer run. If
desired, the light sources (lamps) can be photometered separately, dirt can be removed from the submit-
ted luminaires, or they can be photometered twice, once dirty and once clean, to determine its effects.
This technique has many variations and is the preferred way to write specifications to permit field evalu-
ation of the performance of an installed system. Random sampling at the rate of one per hundred lumi-
naires installed with a minimum of three per installation is a good practice.
Glare
Glare is an especially disturbing influence when viewing a difficult task under low brightness condi-
B-11
tions (such as night). Two types of glare have a critical influence on driver visual performance:
• disability glare (physiological glare), caused by stray light that reduces contrast sensitivity, and thus
produces a loss of visual efficiency
• discomfort glare (psychological glare), ocular discomfort from a bright light source.
Discussions of these two types of glare follow.
Disability Glare. The effect of disability glare is well understood, and relatively easy to quantify. The
scattering of light within the eye produces a luminance that is superimposed on the retinal image,
reducing the contrast level of the image and resulting in a loss of visual efficiency. The summation of the
superimposed luminances due to all of the sources in the visual field is known as the veiling luminance.
Note that the scattering of light within the eye occurs even at relatively low illuminance levels that do not
cause discomfort glare. However, the effect on visual efficiency of a glare source is directly proportional
to the luminous intensity of the source, or luminance of an area, and is inversely proportional to the angle
between the source and the viewing direction.
Discomfort Glare. Discomfort glare is a measure of the level of sensation experienced when a light
source is within the field of view. It cannot be directly measured, and is defined in qualitative rather than
quantitative terms. It is generally accepted that discomfort glare does not have a direct effect on visual
efficiency. However, the long-term effect of discomfort glare is believed to result in fatigue, leading to
potential driver error.
Glare is an especially disturbing influence when viewing a difficult task under conditions of low lumi-
nance, as is typically experienced during nighttime driving.
Quality of Lighting
“Quality of lighting” refers to the relative ability of the available light to provide contrast in the visual
scene in such a manner that people may recognize the cues required for the viewing task. Quality of light-
ing is intimately related to the task. That is, a system that provides high-quality lighting for driving may
be clearly inadequate to meet the needs of an office worker or a tennis player.
The factors involved in producing a high-quality lighting system are interrelated. Care must be taken
in balancing the various parameters of the lighting system to obtain maximum quality. In general, a reduc-
tion in disability glare will improve visibility, and reductions in discomfort glare should improve driver
performance. Glare caused by reflections off the roadway may conceal contrast differences and should
likewise be reduced. However, a change in pavement reflectance, which may reduce reflected glare, will
also change the pavement luminance and may have an adverse effect on the contrast between the road-
way and objects on the road. Each adjustment of a parameter of the lighting system must be evaluated to
obtain the proper compromise.
Light Loss Factor
In most cases, measured values are less than the calculated values of the new, clean lamp and lumi-
naire. Any lighting parameter, including luminance and veiling luminance, can be calculated in terms of
either an initial or maintained value. It is generally assumed (but not always true) that the highest light
level in the life of the installation will be found when the system is first energized and that all effects of
system aging will cause the light level to decrease. For this reason any compensation for system aging
(formerly the “maintenance factor” or MF) is part of the “light loss factor” (LLF).
Light loss factor is the overall factor used to link calculated to measured levels. The lighting design
must incorporate an LLF in all calculations. Light loss factors that change with time after installation may
B-12
Appendix B
be combined into a single multiplying factor for inclusion in calculations. Note that an LLF is composed
of still separate factors, each of which is controlled and evaluated separately. Many of these are controlled
by the selection of equipment (equipment factor) and many others are controlled by planned maintenance
operations (maintenance factor).
LLF is usually considered to be made up of the following:
• reduced output from the light source due to aging—lamp lumen depreciation (LLD).
• reduced output from the luminaire (2–7 percent per year) due to:
– dirt accumulation on lamp
– dirt accumulation on luminaire reflector, refractor, or lens, resulting in reduced transmissivity
– changes in relative light intensity due to dirt accumulation altering the characteristics of lamp,
reflector, and enclosure surfaces
• changes in relative light intensity due to changes in reflector and enclosure characteristics with time
• changes in ballast characteristics (percent regulation) with lamp burn or line voltage time (this fac-
tor is improved with electronic or other more expensive ballast types)
• changes in lamp characteristics with burn time or installation position.
With regard to pavement luminance, the effects of wear, patching, moisture, and repaving must be
added to the above list.
Many of the above factors are complex in nature and are interrelated. Additional information can be
found in other chapters of this guide and in publications of the IESNA, CIE, and others. A few important
points are discussed.
Effect of Lamp Voltage Changes. The high-pressure sodium lamp is the most common light source
used in new roadway lighting installations in the United States, and the relationship between the lamp and
ballast is very complex. It is not usually recognized that these lamps, when used on the most common
types of ballast, produce less light when first installed than after several thousand hours of operation. This
is not the fault of the ballast or lamp but is due to the fact that the lamp voltage rises during life of the
lamp and typical low-cost ballasts do not compensate for this voltage change. This means that for lumi-
naires operating in clean conditions, the light level is likely to increase for some time after initial instal-
lation.
Deterioration of Surface Reflectance. The surface finish of metal reflectors, plastic reflectors, and
plastic enclosures can be scratched and dulled by some maintenance techniques. Sandstorms and expo-
sure to ultraviolet light can also affect the transmission of some plastics. This damage is permanent and
not reversible.
Maintaining Light Levels
In general two approaches can be taken to the question of how to maintain light levels. One is to antic-
ipate the normal maintenance practices and schedules and select an LLF based on estimates of the effects
of those schedules and practices. The other is to select an LLF and then monitor the light level and tailor
the maintenance schedule and practices to what is occurring at the installation. Typical LLF values for
high intensity discharge lighting on highways is 0.6 to 0.65 for high-pressure sodium lamps and 0.45 to
0.5 for metal halide lamps. Designing branch circuits with low voltage drop, the use of electronic ballasts,
and improvements in lamp manufacturing can mitigate LLF.
Coefficient of Pavement Reflectance
In understanding the calculation and use of pavement luminance, it is useful to understand the signif-
icance of the coefficient of reflectance (R) as used in the R-tables. The following explanation is as simple
B-13
and non-technical as possible. This is not a derivation of R.

As a single ray of light striking a surface does so, a portion of the energy is absorbed by the surface, and
the remainder is reflected at a variety of angles. We are interested only in one reflected ray: that which will
reach the eye of the observer. A coefficient of reflectance (R) can be the multiplying factor to be used in
calculating the intensity of the reflected ray as compared to the incident ray. If this concept had been used
in the r-tables, then the numbers would have been very small and would have varied greatly as the angle
of incident light (γ) and as the angle to the line of sight (β) change. Such a relationship can be represented
as
Luminance = Light Intensity (I ) × Reflectance Coefficient (R). (0-5)
The concept used in the “reduced coefficient” R first calculates the incident ray of light as a horizon-
tal illuminance value (Eh):
q(b, g) I (f, g)cos3 g

L⫽ (B-9)
pH 2
and then relates the value of the horizontal illumination at a point on the pavement to the intensity of the
reflected ray directed toward the observer’s eye. By mathematically placing the “cosine cubed gamma”
into the r value (r × cos3 γ × R), it is possible to greatly reduce the spread of the size of the numbers in the
R-table. The numbers are still very small, so they are multiplied by 10,000 before the table is complied.
“Remember from page C-13 that the “reduced coefficient of reflection” R is equal to the directional
reflectance coefficient times cos3γ. Thus the calculation of the pavement luminance, L, may be written as:
I (f, g ⫻ r )
L⫽ ⋅ (in cd/m 2 )
H 2 ⫻ 10, 000
It is a very useful concept to present the coefficient of reflectance in this manner. The correct termi-
nology is “reduced coefficient of reflectance.”
Since the r value is sometimes loosely defined as converting horizontal incident light into luminance,
the erroneous conclusion may be reached that the r value can be used to convert the sum of all horizontal
incident light from several luminaires into luminance at a point defined by the angles of the table. This is
TABLE B-2. Types of Incandescent Lamps
Lamp Type Description Usage
Common Consists of a tungsten filament enclosed in a glass Low initial cost, but low
Incandescent envelope (the bulb) attached to a metal base. The bulb efficacy (less than 22
is evacuated and an inert gas (argon or nitrogen) is lumens per watt) and short
introduced into the bulb for longer life and better lamp life. Typically used
efficiency. For wattages less than 40, the filament in traffic signals.
usually operates in a vacuum, without the inert gas.
A voltage applied to the filament through the base
causes a current to flow through the filament which
causes it to incandesce.
Tungsten Halogen Has a halogen, such as iodine, introduced into Tungsten halogen lamps
(quartz iodide) the lamp envelope. This lengthens lamp life and are not normally used in
improves efficiency. highway lighting. Low
efficency (30 1m/watt)
B-14
Appendix B
not the case. The conversion is limited to the horizontal incident light coming from a single direction
defined by the angles in the table. The position of the point on the pavement is defined and specified by
the observer’s location and angular direction of sight.
Light Sources
The light source (the lamp) is the device that actually converts electrical energy to visible light.
Light sources of interest in highway illumination design fall into two groups—incandescent lamps
and electric discharge lamps. This section covers the various types of incandescent and electric dis-
charge lamps.
Incandescent Lamps
TABLE B-3. Types of Electric Discharge Lamps
Lamp Type Description Usage Notes
Fluorescent Tubular bulb containing mercury. The mercury arc Medium initial cost, long life, high
operating at a low vapor pressure produces ultraviolet efficacy (30–70 lm/w), but light
light. The inside of the bulb is coated with a phosphor, varies with ambient temperature.
and the ultraviolet light striking the phosphor causes Linear light source more difficult to
visible light to be emitted. control.
Induction Same as fluorescent, except not tubular. High efficiency (75 lm/w).
Fluorescent Extremely long 100,000 hour life.
Mercury Vapor Mercury arc in quartz arc tube. Operates at higher Phosphor-coated lamp used for sign
(MV) pressure and temperature than fluorescent. Arc lighting. High initial cost, long life,
produces visible light and some ultraviolet light. high efficacy (30–65 lm/w). Smaller
Far-ultraviolet filtered out by glass envelope. light source than fluorescent.
Clear lamps and phosphor-coated lamps available.
High Pressure Arc in ceramic tube containing sodium and other High initial cost, long life, high
Sodium (HPS) elements. Provides light primarily in yellow efficacy (45–150 lm/w), small light
spectrum but other elements provide light in blue, source.
green, orange and red to improve color rendition.
Requires a starting aid to provide pulse to start arc
stream.
Low Pressure Arc in long tubular glass envelope containing High initial cost, moderately long
Sodium (LPS) sodium only. Light is mono-chromatic yellow with life, high efficacy (145–185 lm/w).
poor color rendering. Light source hard to control
photometrically.
Metal Halide Similar to mercury lamp but contains various metal High initial cost, moderately long
(MH) halides in addition to mercury. Excellent color life, high efficacy (75–125 lm/w).
rendering.
Note: The theoretical maximum efficacy is 683 lm/w. (lm/w = lumens per watt.)
B-15

GL-6

Uploaded by

Copyright:

Available Formats

GL-6

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

GL-6

Uploaded by

Copyright:

Available Formats

cover_7054_AASHTO 10/17/05 11:09 PM Page i

© 2005 by the American Association of State Highway and Transportation Officials.

© 2005 by the American Association of State Highway and Transportation Officials.

ISBN: 1-56051-325-X Publication Code: GL-6

© 2005 by the American Association of State Highway and Transportation Officials.

REGION II: Harold Linnenkohl, Georgia, One-Year Term

REGION III: Frank Busalacchi, Wisconsin, One-Year Term

REGION IV: David Sprynczynatyk, North Dakota, One-Year Term

Chairman: Karl A. Burkett, P.E., Texas

Secretary: Balu Ananthanarayanan, P.E., Wisconsin

Carl Andersen, FHWA

Stanley C. Biddick, P.E., Alabama

Mike DeAngelo, New Jersey

Jon Dickinson, Nevada

Ernest Kim, P.E., Oregon

Susan Lodahl, P.E., Minnesota

Jeff Unick, P.E., Pennsylvania

Chair: Delbert McOmie Chair: Vacant

Associate Member—Bridge, Port, and Toll

Associate Member—Cities and Counties

Chair: Allen D. Biehler Vice Chair: Robert L. Walters

Alabama Don T. Arkle, P.E.

U.S. DOT Member

Associate Member—Bridge, Port and Toll

Associate Member—Bridge, Port and Toll

Associate Member—Cities and Counties

Chapter 1: Introduction ................................................................................................................1

Chapter 2: Master Lighting Plans................................................................................................3

Chapter 3: Techniques of Lighting Design................................................................................13

Luminance Considerations ..............................................................................14

Chapter 4: Techniques of Lighting Design................................................................................31

Chapter 5: Tunnels and Underpasses ........................................................................................33

Chapter 6: Work Zone Lighting and Temporary Roadway Lighting....................................39

Chapter 7: Roundabouts .............................................................................................................41

Chapter 8: Electrical Systems Requirements............................................................................43

Chapter 9: Safety Rest Areas......................................................................................................45

Chapter 10: Roadway Sign Lighting..........................................................................................49

Chapter 11: Maintenance Considerations in Roadway Lighting Design ...............................55

Chapter 12: Sky Glow and Light Trespass ...............................................................................59

Need for Engineering Expertise

Recommended Design Methods

Modern Controls and Master Lighting Plans

2.2 PLAN DEVELOPMENT

TABLE 2-1. Master Lighting Plan Development Process

Step/Action More Information

Groups Having Concerns About Lighting

2.3 GOALS OF A MASTER LIGHTING PLAN

A master lighting plan should have five major goals:

Environmentally Judicious Use of Resources

TABLE 2-2. Resources for Lighting Engineering Guidelines

Organization Internet Address

The International Commission on Illumination (CIE) http://www.cie.co.at/cie

Tourists, Businesses, and Nighttime Activities

2.4 LIGHTING CURFEWS

Reasons for Curfews

Considerations Before Implementation

2.5 ELECTRICAL AND LIGHTING MANAGEMENT SYSTEMS

monitoring and control.

The benefits of ELMS are outlined in the following table.

Benefits of ELMS Standardization