NIGHT VISIBILITY UNDER DIFFERENT SYSTEMS OF FIXED ROADWAY LIGHTING: A PROGRESS REPORT

The visibility of objects on the roadway at night has been studied under different conditions of fixed lighting and pavement reflectance in a 15:1-scale model of a divided urban freeway. Visibility was measured in accordance with the quantitative methods described by the International Commission on Illumination, involving psychophysical measurements of equivalent contrast and physical measurements of disability glare and luminance. Initially, three realistic targets were studied at 20-ft(6.1-m) intervals throughout the luminaire cycle down the center of each of eight lanes, under six layouts of model luminaires. Average results indicated that visibility did not increase so much with increases in illuminance as was expected. Also, when the spacing of luminaires was wide, variability in visibility was great. These results were attributed to an anticamouflage effect of roadway luminance nonuniformities. Subsequent research with identical targets of different reflectances confirmed this and indicated that the visibility potential of different pavements and lighting systems should be based on visibility measures of targets of the normal reflectances weighted in proportion to the frequency of their occurrence. The quantity of data needed for this purpose implies need for a physical correlate of target visibility. Target visibility can be predicted reasonably well from roadway luminance and the arithmetic average of the local contrasts of different parts of a target with adjacent portions of the roadway background. It now seems possible to compute indexes of visibility potential for different roadway environments. Further work is needed to asses transient adaptational effect and individual differences in visibility threshold.