A methodology for locating the maximum vertical illuminance in street lighting.

1 INTRODUCTION

During the dark hours, drivers are faced with the complex tasks of maneuvering the car, seeing other cars, identifying the road and identifying objects and Pedestrians. Therefore, reacting to the presence of pedestrian is only one of the many tasks that the driver faces. A report by the US Federal Highway Administration indicates that 62 percent of pedestrian fatalities happened at night. The fatal crashes peak in the evening hours between 5-11PM. The study also indicates that 60 percent of pedestrian accidents occur at places other than intersections (Freedman 1975). Sullivan and Flanagon (2007) found that "for equal exposure, the risk of a pedestrian fatal crash in darkness is on average almost seven times greater than in daylight."

The objective of this study is to develop a methodology for finding the maximum vertical illuminance 1.5 m above pavement level along streets. The location of cross walks using the vertical illuminance along the street as it's metric is one application of such a methodology. It is to be noted that, a higher vertical illuminance may not necessarily mean better visibility. That is because visibility is a function of many other factors listed below. The vertical illuminance is a convenient metric to use in the design process because of its simplicity and ability to be calculated using many commonly available software packages. Whereas, the Illuminating Engineering Society of North America (IESNA) does not have cross walk recommendation, RP-8 (IESNA 1999) does refer to pedestrian walkways using vertical illuminance 1.5 m above pavement level.

1.1 PRINCIPLES OF NIGHT TIME VISIBLITY

During night time, pedestrians are very hard to see. Unfortunately pedestrians falsely, assume that drivers can actually see them as a result they unknowingly place themselves in danger (Tirell el al 2004).

The human visual system can detect objects because of the following main reasons:

1- Luminous Contrast: The contrast can be positive which occurs when the object is brighter than the surround, or negative which is when the object is darker than its background as shown in Fig. 1.

[FIGURE 1 OMITTED]

2- Chromatic contrast which means that the color of the object is different from its background.

3- Adaptation luminance which refers to the state of the adaptation of the eye. In general, the eye adapts to the brightest spot in the visual field. If there is a very bright light source within the visual filed of the eye, the source can either cause blinding glare which causes the eye not to be able to see other objects, or it can cause discomfort glare which causes eye fatigue.

4- The size of the object relative to the sight distance. Small object that are far away are harder to see than objects of the same size but are closer to the eye.

5- Position of the object within the visual field. Is the object directly in front of us, or behind us or on the side?

6- Luminous flux, which causes contrast. In other words, we need light to see

7- Objects movement. The eye see objects that are moving better than objects that are still.

In addition to the above main factors, there are other elements of visibility such as:

a. Expectation; we see objects that we expect faster than unexpected objects.

b. Time available to see the objects. The longer the time we have the better we see the objects.

c. Age of the observer and his/her vision status

d. Obstructions that are present between the line of sight and the object.

Should a driver see a pedestrian crossing the road, the driver needs some time to react to this perception. This time is a function of a lot of factors such as expectations, age, focus, and so on In general the reaction time of 2.5 second during which the car would have traveled some distance. In addition to this distance there is the distance that it takes the car to reach complete stop. Total Stopping Distance (TSD) for different speeds assuming a perception and response time (P & RT) of 2.5 seconds are listed in Table 1.

1.2 EFFECT OF STREET LIGHTING ON PEDESTRIAN VISIBILITY

While automotive headlight objective is to assist the drivers in locating and seeing objects along their way, the reach and the spread of headlight is not enough for locating pedestrians. Pedestrians could be on the side away from the headlight spread or they could be far from headlights reach. Studies done at the University of Michigan Transportation institute indicate that low beam visibility distances to unexpected low contrast objects are shorter than the stopping distances for speed above 70km/hour (Sivak and others, 2006).

Sivak (Sivak and others, 2006) showed that the low beam vertical illuminance of 3 Lux, 0.25m above ground reached 38m on the left side, 65m on the center and 100m on the right edge of the street. See Fig. 2.

Studies have shown that drivers using low beam in an un-illuminated street cannot recognize an unexpected dark clad pedestrian until he or she approached to a distance of 15-45m (Leibowitz and others, 1998). Under most conditions the visibility distance for pedestrians at night is much shorter than the total stopping distance (Leibowitz 1998).

Therefore, installing street lights greatly improves night time visibility. First, the vertical illuminance on pedestrians increases, second the glare from oncoming headlights is reduced (Bacelear 2004), third, the ability to detect low contrast objects is higher at higher road luminance levels (Eloholma 2004), furthermore, streetlights will illuminate objects that are outside the reach and spread of car headlights.

1.3 EFFECT OF LIGHT SOURCES ON PEDESTRIAN VISIBILITY

The predominant light sources currently used in street lighting are either metal halide or high pressure sodium with mercury vapor and low pressure sodium fading away from been used. The effectiveness of the type and color of light source to improve night time visibility of pedestrians has been the subject of many recent studies (Akashi and others, 2007, Lin and others, 2006, Fotios and others, 2005) to name a few.

However, many studies have showed that peripheral vision is improved using metal halide lamps as opposed to high pressure sodium lamps. A detailed overview of those studies is made by Fotios and others, (2005). The visual field does not seem to be affected by the spectrum of the source (Lin and others, 2004).

2 ILLUMINATION OF CROSS WALKS

Typically street lighting is designed based on the road class (Local, Major, Collector) and the level of pedestrian activities (IESNA RP-8,1999). The recommended practice RP-8 does not have any standards for cross walk illumination requirement. RP-8 does, however, outline illuminance levels needed for walkways and bikeways. The CIE technical report 115 (CIE 1995) specifies average horizontal illuminance for crosswalks values ranging from 20-1.5 Lux based on type of roads. A modified visibility model based on Adrian's model was presented by Ising (2008) along with a review of various recommended Visibility Levels.

[FIGURE 2 OMITTED]

The US Federal Highway Administration (FHA) developed a study titled 'Fixed Illumination for Pedestrian Protection; User's Manual" (Freedman et. al 1975) which contained recommendation for crosswalk lighting. The manual recommended providing at least 75 Lux average horizontal illuminance along the cross walk; provide color contrast from the rest of the street to clearly designate the crosswalk and, provide a uniformity of [E.sub.avg]/[E.sub.min] 4:1. A more recent study by FHA (Gibbons and others, 2008) looked at the location of luminaires at crosswalks and recommended 20 lx vertical illuminance level 1.5m above street level. A contour of the vertical illuminance on a plan view of one way street was shown by Gibbons el al for a 250w HPS Flat lens cobra-head 10 m and 8.5 m high.

3 METHODOLOGY

A typical street is designed to the illuminance levels recommended by IESNA RP-8 assuming high pedestrian activity levels. To accomplish those levels, a street lighting engineer has many variables such as:

1. lamp wattage and type.

2. light fixture optical class (example medium, type III, cutoff).

3. mounting height.

4. spacing and.

5. fixture layout such as staggered, median mounted Fig. 3.

A roadway lighting optimizer (Lighting Analysts 2006) was used to get the spacing needed for various fixture layouts. This spacing was then used to get more detailed vertical illuminance levels. Vertical grids 2 m apart were divided into 3 grids. One grid on the side walk, another one was on top of the street, and the third one was on top of the median. All grids were designed to measure vertical illuminance 1.5 m above street level. This is illustrated in Fig. 4. This configuration simulates pedestrians on the side or crossing the street or waiting on the median. The vertical illuminance [E.sub.v] facing oncoming traffic was calculated for those grids, then the minimum value [E.sub.v(min)] was found for each grid. The locations between two poles that provide maximum [E.sub.v(min)] were found. The distance to one pole divided by the spacing between the poles (D/S) is the ratio used to determine the location of the maximum [E.sub.v(min)].

Lanes are assumed to be 3.5m wide on R3 pavement with the poles having a 2m bracket and 1.5m setback, while the light loss factor (LLF) is assumed to be 0.8. Median mounted fixtures were placed in the center of the median and had 1 m bracket. For the purpose of this study, light from automotive headlights, were not considered in the calculations of vertical illuminance.

4 RESULTS

Figure 5 illustrates one situation for a 250w HPS lamp, type III medium full cutoff fixture with a mounting height of 12 m, and a longitudinal spacing of 95m between poles placed on both sides (Layout f). The peak minimum illuminance indicates the location between the poles that provides maximum [E.sub.v(min)]. In this case, the location that provides the highest vertical illuminance 1.5m above the street at a distance from one pole (Farther pole relative to the direction of traffic flow) that is 0.93 of the spacing between the poles. That is maximum [E.sub.v(min)] for the street (PED) grid is at the same location as the side and median grids. Because traffic flow on the other side of the street is in the other direction, the vertical illuminance trend is opposite that of the first side of the street. A plan view shown in Fig. 6 illustrates the situation.

When a cross walk needs to be designed to cross two way street with a median, the cross walk can be on one lateral axis or it can be a on two different axes. The vertical illuminance on two way street changes direction to face the incoming traffic. Therefore if the location of the maximum [E.sub.v(min)] on one side of the street is not the same as the other side, cross walk that are on one axis may not provide the maximum [E.sub.v(min)] Throughout the length of the crosswalk. Figure 7 and Fig. 8 show an example of this situation. Because of nonsymmetry between the two ways, the max [E.sub.v(min)] on one side will not occur at the same point laterally as the other.

Semi cutoff fixtures provided fewer locations with [E.sub.v(min)] = 0 as shown in Fig. 9. This result does not mean that semi cutoff fixtures provide better visibility. Semi cutoff fixtures produce more glare which can reduce visibility. With many municipalities and counties implementing dark sky compliant fixtures, semi cutoff fixtures are being phased out.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

Figure 10 shows a summary of the distance over spacing (D/S) ratios that provide maximum [E.sub.v(min)] for various arrangements. The distance form one pole relative to the spacing between poles ranged from 0.6 to 0.9. The parameter D in this article is defined as the distance from the farther pole as seen from the direction of traffic flow to the grid.

The locations and configurations that result in max [E.sub.v(min)] on, PED grid which is above street and on Median grid are illustrated in Table 2 and Table 3 respectively. Because the spacing is actually between two poles that are on the same side of the street, for staggered layout, two local maxima are found.

It is to be noted that a configuration that gives very high [E.sub.v(min)] at some locations while providing [E.sub.v(min)] < 0.2 FC at other locations is not recommended for streets that do not force pedestrians to use the cross walk. Streets that force pedestrians to use cross walk have fences and/or obstructions that physically prevent pedestrians from crossing the street at any location other than a specified location. In streets that do not force pedestrians to use crosswalks, the configuration that provides maximum flexibility should be chosen. This configuration is typically the staggered layout with lamps at high mounting height.

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

5 CONCLUSION

This article developed a method to find the location for pedestrian crosswalks which provide the maximum vertical illuminance. This helps city planners in locating pedestrian crosswalks.

It was found that the minimum vertical illuminance [E.sub.v(min)] 1.5m above ground can be zero despite the fact that the horizontal illuminance on the street satisfy the IESNA recommendations. For two way streets, the vertical illuminance [E.sub.v(min)] can be maximum on one side while it is minimum on the same lateral axis but the other side of the street. So, pedestrians might be seen in positive contrast on side of the street and in negative contrast on the other side. This shows the difficulty of placing the crosswalk on one lateral axis.

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

[FIGURE 9 OMITTED]

The distance from the farther pole to the point of the maximum [E.sub.v(min)] divided by the spacing between the poles (D/S) was used. The values of (D/S) for different streetlighting layout were found to be between 0.6 and 0.9.

[FIGURE 10 OMITTED]

This study concerned itself with minimum vertical illuminance 1.5m above street level. Future studies need to investigate the visibility of pedestrians and should include the effect of street lighting on pedestrian contrast with background. Furthermore, the interaction between automotive headlights and the street lighting to see pedestrians need to be studied. To provide minimum vertical illuminance greater than 3 FC would generally requires supplemental poles to illuminate the cross walk.

A article by the author (Saraiji 2009) shows a more comprehensive analysis of various streetlight configurations. Future studies would include contrast analysis of pedestrians relative to the street and surroundings. Considerations for eye adaptations, pedestrian clothing, movement, and so on are all factors that influence visibility and are to be included in future studies.

REFERENCES

American Association of State Highway and Transaction Officials. 2001. A policy on geometric design: Highways and street. 4th Edition.

Akashi Y, Rea M, Bullough J. 2007. Driver decision making in response to peripheral moving targets under mesopic levels. Lighting Res. Technol. 39,1 (2007) pp. 53-76.

Australian/New Zealand Standard. 1999. Road lighting part 3.1: Pedestrian area (Category P) lighting-performance and installation design requirements. As/NZS 1158.3. 1:1999.

Bacelar A. 2004. The contribution of vehicle lights in uran and peripheral urban environments. Light Res Tech. 36(1): 69-78.

[CIE] Comission Internationale de L'Eclairage. 1995. Recommendations for the lighting of roads for motor and pedestrian traffic. CIE 115. ISBN 3 900 734 59 3.

Ekrias A. 2007. Road lighting and headlights: Luminance measurements and automobile lighting solutions. Building Environ. 43(4):530-536.

Eloholma M. 2005. Mesopic models-from brightness matching to visual performance in night-time driving: a review. Light Res Tech. 36(2):155-175.

Fotios S, Cheal C, Boyce PR. 2005. Light source spectrum, brightness perception and visual performance in pedestrian environments: a review. Light Res Tech. 37(4):271-294.

Freedman. 1975. Fixed Illumination for Pedestrian Protection; User's Manual. FHWA-RD-76-9.

Gibbons RB, Edwards C, Williams B, Andersen C. 2008. Informational report on lighting design for midblock crosswalks. FHWA-HRT-08-053.

Guler O, Onaygil S. 2003. The effect of luminance uniformity on visibility level in road lighting. Light Res Tech. 35(3):199-215.

[IESNA] Illuminating Engineering Society of North America. 1999. RP-8, Recommended Practice for Roadway Lighting

Ising K. 2008, Threshold visibility levels required for nighttime pedestrian detection in a modified Adrian/CIE visibility model. Leukos. 5(1):63-75.

Knoblauch R, Raymond P. 2000. The effect of crosswalk marking on vehicles speeds in Maryland, Virginia, and Arizona. U.S. Department of Transportation. Report No.FHWA-RD00-101.

Leden L, Grader P, Johansson C. 2006. Safe pedestrian crossing for children and elderly. Accident Analysis and Prevention. 38(2):289-294.

Leibowitz H, Owens D, Tyrrell R. 1998. The assured clear distance ahead rule: implications for nighttime traffic safety and the law. Accidents. Analysis. and Prevention. 30(1): 93-99.

Lighting Analysts. 2006. AGI32 version 1.95 revision 2.

Lin Y, Chen W, Chen D, Shao H. 2004. The effect of spectrum on visual field in road lighting. Building and Environ. 39(4):433-439.

Lin Y, Chen D, Chen W. 2006. The significance of mesopic visual performance and its use in developing a mesopic system. Building and Environment. 41(2):117-125.

Nitzbing M, Knoblauch R. 2001. An Evaluation of High-Visibility Crosswalk Treatment-Clearwater, Florida. U.S. Department of Transportation. Report No.FHWA-RD-00-105.

Saraiji R. 2009. Vertical Illuminance Based Crosswalk Illumination. Leukos 6(2):153-167.

Schoettle B, Sivak M, Flannagan MJ, Kosmatk W. 2004. A market-weighted description of low-beam headlighting patterns in the U.S.: 2004. The University of Michigan Transportation Research Institute. Report No. UMTIR-2004-23.

Sekuler R, Blake R. 1990. Perception. 2nd Edition. McGraw-Hill. 519p.

Sivak M, Schoettle B, Minoda T, Flannagan. 2006. Potential visibility gains on straight and curved roads from proportional increases in low-beam headlamp intensities. Light Res Tech. 38(3):259-266.

Sullivan JM, Flannagan MJ. 2002. The role of ambient light level in fatal crashes: inferences from daylight saving time transactions. Accident Analysis and Prevention. 34(4):487-498.

Sullivan JM, Flannagan MJ. 2007. Determination of the potential safety benefit of improved lighting in three pedestrian crash scenarios. Accident Analysis and Prevention 39(3): 638-647.

Tyrrell R, Wood J, Carberry T. 2004. On-road measures of pedestrian's estimates of their own nighttime conspicuity. J Safety Res. 35(5): 483-490.

Nomenclature

[E.sub.v(min)]: Minimum vertical illuminance facing oncoming traffic 1.5 m above pavement level.

PED: Pedestrian calculation grid of [E.sub.v(min)] over the street and 1.5 m above pavement level.

D: Distance between one light pole (farther relative to the direction of traffic) and a calculation grid of [E.sub.v(min)].

S: Spacing between poles

MH: Mounting height of lamp

[E.sub.h]: Horizontal street illumination for a calculation grid between two street poles.

Riad Saraiji, PhD

Architectural Engineering Department, United Arab Emirates University, P.O. Box 17555, El Ain, United Arab Emirates, e-mail [e-mail], Phone: [phone]

doi: 10.1582/LEUKOS.2009.06.02004

TABLE 1. Sight Distance Required for
Design Speed (ASHTO 2001)

Design Speed (KmPH)   2-Lane Rd (m)   4-Lane Rd (m)

        40                  75              85
        50                  95             110
        60                 115             130
        70                 135             150
        80                 150             175
        90                 170             195
       100                 190             215
       110                 210             240

TABLE 2. The Location Between Light Poles that Provide the Maximum
Vertical Illuminance on PED Grid

  Light       Mounting      Configuration     Spacing (m)
 Fixture       Height                           Based on
(Type III)    (meters)         3 Lanes        [E.sub.avg]

 150w HPS        12       One side                 27
 150w HPS        12       Staggered                52
 150w HPS        12       Both sides               52
 150w HPS        12       2 rows median            34
 150w HPS        12       4 rows opposite          70
 150w HPS        12       4 rows staggered         69
 250w HPS        12       One side                 39
 250w HPS        12       Staggered                77
 250w HPS        12       Both sides               78
 250w HPS        12       2 rows median            58
 250w HPS        12       4 rows opposite          95
 250w HPS        12       4 rows staggered        104
 150w HPS        10       One side                 31
 150w HPS        10       Staggered                48
 150w HPS        10       Both sides               60
 150w HPS        10       2 rows median            38
 150w HPS        10       4 rows opposite          58
 150w HPS        10       4 rows staggered         69
 250w HPS        10       One side                 45
 250w HPS        10       Staggered                89
 250w HPS        10       Both sides               59
 250w HPS        10       2 rows median            58
 250w HPS        10       4 rows opposite          68
 250w HPS        10       4 rows staggered        116

  Light      max [E.sub.[alpha](min)]
 Fixture
(Type III)         Dist/Spacing

 150w HPS            0.63
 150w HPS    0.35            0.85
 150w HPS            0.85
 150w HPS            0.59
 150w HPS            0.86
 150w HPS    0.29             0.8
 250w HPS            0.77
 250w HPS    0.34            0.36
 250w HPS             0.9
 250w HPS            0.84
 250w HPS            0.91
 250w HPS    0.29            0.42
 150w HPS            0.58
 150w HPS    0.25            0.75
 150w HPS             0.7
 150w HPS            0.63
 150w HPS            0.86
 150w HPS    0.29             0.8
 250w HPS            0.84
 250w HPS    0.43            0.31
 250w HPS            0.88
 250w HPS            0.86
 250w HPS            0.88
 250w HPS    0.43            0.33

TABLE 3. The Location Between Light Poles that Provide Maximum Ev(min)
on the Median Grid

  Light       Mounting      Configuration     Spacing (m)
 Fixture       Height                          Based on
(Type III)    (meters)         3 Lanes           Eavg

 150w HPS        12       One side                27
 150w HPS        12       Staggered               52
 150w HPS        12       Both sides              52
 150w HPS        12       2 rows median           34
 150w HPS        12       4 rows opposite         70
 150w HPS        12       4 rows staggered        69
 250w HPS        12       One side                39
 250w HPS        12       Staggered               77
 250w HPS        12       Both sides              78
 250w HPS        12       2 rows median           58
 250w HPS        12       4 rows opposite         95
 250w HPS        12       4 rows staggered        104
 400w HPS        12       One side                74
 400w HPS        12       Staggered               141
 400w HPS        12       Both sides              84
 400w HPS        12       2 rows median           75
 400w HPS        12       4 rows opposite         87
 400w HPS        12       4 rows staggered        137
 150w HPS        10       One side                31
 150w HPS        10       Staggered               48
 150w HPS        10       Both sides              60
 150w HPS        10       2 rows median           38
 150w HPS        10       4 rows opposite         58
 150w HPS        10       4 rows staggered        69
 250w HPS        10       One side                45
 250w HPS        10       Staggered               89
 250w HPS        10       Both sides              59
 250w HPS        10       2 rows median           58
 250w HPS        10       4 rows opposite         68
 250w HPS        10       4 rows staggered        116

  Light             Max Ev(min)      max Ev Min
 Fixture                             on Median
(Type III)         Dist/Spacing         (FC)

 150w HPS            0.63               0.5
 150w HPS    0.35            0.85       1
 150w HPS            0.85               1.1
 150w HPS            0.59               1
 150w HPS            0.86               2.2
 150w HPS    0.29             0.8       1.9
 250w HPS            0.77               0.6
 250w HPS    0.34            0.36       0.8
 250w HPS             0.9               1.2
 250w HPS            0.84               1.6
 250w HPS            0.91               2.9
 250w HPS    0.29            0.42       1.6
 400w HPS            0.81               1.1
 400w HPS    0.91             0.4       2.8
 400w HPS            0.83               3.7
 400w HPS            0.83               8.3
 400w HPS            0.88               8.3
 400w HPS    0.91            0.41       8.5
 150w HPS            0.58               0.7
 150w HPS    0.25            0.75       1.2
 150w HPS             0.7               0.9
 150w HPS            0.63               1.5
 150w HPS            0.86               2.6
 150w HPS    0.29             0.8       2.3
 250w HPS            0.84               0.7
 250w HPS    0.43            0.31       1
 250w HPS            0.88               1.5
 250w HPS            0.86               2.3
 250w HPS            0.88               3.6
 250w HPS    0.43            0.33       2.7