Discomfort Glare Metrics: Elusive, but not Illusive: A companion piece to The Elusive Discomfort Glare Metric

Editorial Disclaimer
The views expressed in articles published on FIRES do not necessarily reflect those of IES or represent endorsement by the IES.

By John D. Bullough, Ph.D.
Lighting Research Center, Rensselaer Polytechnic Institute

Naomi Miller’s essay “The Elusive Discomfort Glare Metric” highlights a number of questions within the lighting community about the causes of (and, hopefully, the cures for) discomfort glare, as well as some recent progress from the International Commission on Illumination (CIE) toward practical metrics that address solid-state lighting (SSL) systems. Playing on the title of her essay, this comment hopes to demonstrate that the path to appropriate metrics for discomfort glare may require expert navigational skills (elusive), but ultimately it is not a path to nowhere (illusive).

The workaround for the unified glare rating (UGR) published by CIE earlier this year, which uses only the high-luminance (>500 cd/m²) portions of a luminaire in estimating its potential for creating discomfort glare is a step in the right direction because higher maximum luminances will certainly result in greater visual discomfort, holding all other factors equal. And while they may not be readily apparent, preliminary answers to the six questions in Naomi Miller’s essay exist, but they can be hard to find because the literature on discomfort glare is spread across many different disciplines, including not only lighting but vision, transportation and optics. One synthesis on the prediction of discomfort glare was published in the proceedings of a recent automotive lighting symposium (Bullough 2017).

1. Glare source luminance: From studies of glare from automotive lighting (Rosenhahn and Lampen 2004; Bullough 2011), a size of 0.2 to 0.3 degrees seems to be appropriate for the size of the maximum luminance as it affects discomfort glare. If the size is smaller, only the illuminance from the source impacts discomfort glare (Bullough and Sweater Hickcox 2012; see Fig. 1). This is convenient because commercially available luminance meters with an aperture size of 0.3 degrees exist.

2. Size of glare source: This can be avoided by using the illuminance from the glare source and the maximum luminance (Bullough et al. 2011; Bullough and Sweater Hickcox, 2012; see also Fig. 1). If the glare source is smaller than 0.3 degrees in diameter, only the illuminance is needed.
    
3. Background luminance: Work from Schmidt-Clausen and Bindels (1974) showed that if two or more point sources were no more than 3 degrees apart, their discomfort glare effects could be predicted by assuming they were a single source producing the sum of the individual sources’ illuminances at observers’ eyes. Bullough et al. (2008) showed that when glare sources were at least 9 degrees apart, their discomfort impacts were independent of each other. Therefore, the background for a given glare source should probably have a radius of somewhere between 3 and 9 degrees. Since this background can still have a widely varying luminance distribution, the illuminance from this angular region can stand in as a surrogate for the immediate background luminance. However, the light from all other angles beyond this approximately 6-degree radius still influences discomfort (Bullough et al. 2008).
    
4. Adaptation luminance: This issue is not entirely independent of #3. However, visual adaptation is largely local (as demonstrated in the study by Akashi et al. 2007), so there isn’t a single adaptation level for the eyes overall in any given situation.
    
5. Angular deviation: Discomfort glare definitely decreases when the glare source is not close to the line of sight. However, discomfort as experienced in a given lighting installation is not necessarily dependent only upon a particular viewing geometry (Bullough et al., 2003). People look around—especially under nighttime lighting, and including looking more or less directly at offending light sources—and their impressions of how glaring an installation is will be made based on that “worst case” scenario. Many discomfort glare models (especially for outdoor lighting) predict infinite glare when the glare source is directly along the line of sight, which is not helpful, because even when glare is considered unbearable it rarely achieves infinity. The discomfort glare model from the Outdoor Site-Lighting Performance (OSP) system for estimating light pollution (sky glow, light trespass and discomfort glare) assumes direct view of the offending luminaire and seems to work well not only for nighttime conditions (Sammarco et al. 2011; Tyukhova and Waters 2018) but also for interior lighting conditions (Mou et al. 2017).
    
6. Number of glare sources: As mentioned under #3, the number of glare sources can be addressed by determining how many of them fall within approximately 6 degrees of each other. When they are farther apart they can be assessed independently (Bullough et al. 2008).

Though not included in this essay, the spectral distribution also impacts discomfort glare, as many people who have experienced so-called “xenon” headlights while driving at night will loudly attest. In short, greater short-wavelength energy or higher correlated color temperature (CCT) increases discomfort glare, but not disability glare (Rea and Bullough 2018; Bullough et al. 2019), even if luminous intensity is held constant (Bullough 2009; Sweater Hickcox et al. 2013; Bullough and Liu 2019; see Fig. 2). Nonetheless, the spectral distribution of a “white” glare source, likely to be experienced under most lighting conditions, is a relatively unimportant factor.

Importantly, however, even if all of these findings were incorporated into a single, less elusive discomfort glare metric, precise predictions of discomfort glare will never be possible (Bullough 2017), because sensations of discomfort are also impacted by psychological factors such as aesthetics – beautiful glare sources are less “glaring” than ugly ones—and task difficulty—oncoming headlights are more “glaring” on unfamiliar roads than on the familiar commute home—or when visual tasks have low contrast (Sivak et al. 1991; Van Derlofske et al. 2004; see Fig. 3). Still, such a metric or metrics will be useful for relative comparisons between and among different configurations.

References
Akashi Y, Rea MS, Bullough JD. 2007. Driver decision making in response to peripheral moving targets under mesopic light levels. Lighting Research and Technology 39: 53.

Bullough JD. 2009. Spectral sensitivity for extrafoveal discomfort glare. Journal of Modern Optics 56: 1518-1522.

Bullough JD. 2011. Luminance versus luminous intensity as a metric for discomfort glare. SAE Technical Paper 2011-01-0111.

Bullough JD. 2017. Developing a better understanding of discomfort glare: Cause and effect. 12^th International Symposium on Automotive Lighting (pp. 705-714), Darmstadt, Germany, September 25-27.

Bullough JD, Brons JA, Qi R, Rea MS. 2008. Predicting discomfort glare from outdoor lighting installations. Lighting Research and Technology 40: 225.

Bullough JD, Liu Y. 2019. Response to white light-emitting diode aviation signal lights varying in correlated color temperature. Transportation Research Record 2673: 667.

Bullough JD, Skinner NP, Rea MS. 2019. Impacts of flashing emergency lights and vehicle-mounted illumination on driver visibility and glare. SAE Technical Paper 2019-01-0847.

Bullough JD, Sweater Hickcox K. 2012. Interactions among light source luminance, illuminance and size on discomfort glare. SAE Technical Paper 2012-01-0269.

Bullough JD, Sweater Hickcox K, Narendran N. 2012. ASSIST Recommends: A Method for Estimating Discomfort Glare from Exterior Lighting Systems. Troy, NY: Lighting Research Center.

Bullough JD, Van Derlofske J, Fay CR, Dee P. 2003. Discomfort glare from headlamps: Interactions among spectrum, control of gaze and background light level. SAE Technical Paper 2003-01-0296.

Mou X, Freyssinier JP, Narendran N, Bullough JD. 2017. Preliminary evaluation of discomfort glare from OLED and edge-lit LED lighting panels. Journal of Biomedical Optics 22: 055004.

Rea MS, Bullough JD. 2018. Two types of glare: Two visual channels. Scandinavian Journal of Optometry and Visual Science 11: 8.

Rosenhahn EO, Lampen M. 2004. New investigation of the subjective glare effect of projection type headlamps. SAE Technical Paper 2004-01-1281.

Sammarco JJ, Mayton AG, Lutz T, Gallagher S. 2011. Discomfort glare comparison for various LED cap lamps. IEEE Transactions on Industry Applications 47: 1168.

Schmidt-Clausen HJ, Bindels JTH. 1974. Assessment of discomfort glare in motor vehicle lighting. Lighting Research and Technology 6: 79.

Sivak M, Flannagan M, Ensing M, Simmons CJ. 1991. Discomfort glare is task-dependent. International Journal of Vehicle Design 12: 152.

Sweater Hickcox K, Narendran N, Bullough JD, Freyssinier JP. 2013. Effect of different coloured luminous surrounds on LED discomfort glare perception. Lighting Research and Technology 45: 464.

Tyukhova Y, Waters CE. 2018. Discomfort glare from small, high-luminance light sources when viewed against a dark surround. Leukos 14: 215.

Van Derlofske J, Bullough JD, Dee P, Chen J, Akashi Y. 2004. Headlamp parameters and glare. SAE Technical Paper 2004-01-1280.