Research investigates potential hazards when driving in fog

A recent experiment, led by Professor Steve Fotios, aimed to investigate how well we can see potential hazards when driving, after dark, in fog.

To drive safely we need to be able to acquire and process visual information effectively, including speed control, lane control, and looking out for potential hazards. Driving is a complex task and if vision is impaired, the ability to do it safely is inhibited.

While driving after dark we use head lighting and road lighting to offset the visual impairment. However, the presence of fog reduces visibility of distant objects, from about 1000 metres in clear conditions to around 600 metres with thin fog and 75m in thick fog.


Participant ViewFog also scatters light which reduces the contrast of an object against its background, making it more difficult to see. This scattering means that too much light, and from the wrong location, can impair rather than improve the visibility of objects.

In order to test the potential hazards when driving in fog, researchers at the Sheffield School of Architecture built a 1/10 scale model of a three-lane carriageway. Test participants, recruited from a range of ages and balanced across genders, observed this scene from the viewpoint of a driver in the middle lane. Within the chamber the researchers could change the amount (luminance) and spectrum (S/P ratio) of light and they could control the presence and density of fog.

‘Drivers’ were required to detect two events: an object suddenly appearing on the road surface and a car ahead moving into the driver’s lane. To see how the performance was affected by changes in lighting and fog, researchers recorded the reaction time to detection.


Mean reaction times for detection of car by age group, fog condition and light conditionThe results suggested that when driving in thick fog, there is an advantage in road lighting of luminance 1.0 cd/m2 compared with 0.1 cd/m2. We see objects by the light they reflect and luminance tells us how much light is reflected. Typically, 0.1 cd/m2 would be dim road lighting and 1.0 cd/m2 would be relatively bright road lighting.

Drivers showed a quicker reaction time to the obstacles in thick fog with the lighting of higher S/P ratio. This suggests a benefit of using the higher S/P ratio which will appear whiter than the orange sodium lighting traditionally used in the UK.

Funded by Highways England, the University of Sheffield School of Architecture’s Lighting Research Group undertook the experiment in collaboration with Arup. The results will help Highways England develop their strategy for road lighting, for example, where, when and how much lighting to use.

Sheffield staff: Professor Steve Fotios, Dr Chris Cheal, Dr Jim Uttley, Scott Fox