How to Light

How to Light: ASK THE DOCTOR: Mesopic lighting

Dr Gareth Jones is here to answer your technical questions. This month, he enters the twilight zone of mesopic lighting

Q: What is mesopic lighting and why is it important?

To explain mesopic lighting we first need to understand the way the human eye responds to light. The eye is a visible light detector, and detection happens in the retina. The eye’s response to different wavelengths (which we see as colours) is based upon two types of photoreceptors in the retina – rods and cones.

The rods are the most numerous of the photoreceptors – we all have about 120 million of them in each eye. They are responsible for our dark-adapted night-time, or scotopic, vision. The rods are incredibly efficient photoreceptors – more than 1,000 times as sensitive as the cones. But they are not sensitive to colour. They also adapt to light conditions much more slowly than the cones, so optimum dark-adapted vision is obtained only after a considerable period of darkness, say half an hour or longer. We all know this from the way we become ‘used to the dark’ and can see more the longer we spend in the dark.

As for cones, there are approximately seven million in the eye and they allow us to perceive colours. Sixty-four per cent of them are red cones (responding to red light), 32 per cent green and two per cent blue. The green and red cones are concentrated in the fovea – the area of the retina used for central vision. The blue cones have the highest sensitivity and most of them are found outside the fovea.

The colour-sensitive cones are responsible for our photopic (daytime) vision and the rods for our scotopic (night-time) vision.

The difference between the photopic and scotopic responses in terms of colour can be seen in the familiar response curve (see chart).

In lighting we usually use the photopic response function as the basis of everyday measurements of illuminance. But this only applies when the eye is adapted for photopic daytime vision. The photopic lux measurement is a poor predictor of the illuminance in low-light conditions because the eye will not respond according to the photopic response.

Of course, in real life the low light levels at which the eye’s scotopic response is dominant are not associated with visual tasks. But there is a third region of light levels that is not low enough to be dark or bright enough to be light, and this mixed region is known as the mesopic region.

Between two stools

Mesopic lighting is the in-between state where both the rods and cones are being used at light levels between darkness and daytime illumination. It becomes important, for example, for road lighting at night. The amount of lighting used to enable a visual task such as reading or driving in the mesopic region will be different to that in the scotopic or photopic vision scenarios.

Studies are being carried out to work out the best colour mix and illumination levels for our mesopic vision, but keeping energy consumption to a minimum. The scotopic-photopic ratio is becoming an important parameter for interpreting the efficacy of lighting at particular illumination levels.

This is an exciting area of current research, particularly for roadway and pathway lighting
and lighting in towns and cities. It will be important to ensure the best lighting can be installed for all applications with the minimum energy consumption to meet increasingly stringent energy-efficiency targets.


Q: What is a colour-rendering index and why does it matter?

The colour-rendering index or CRI of a light source is a measure of its ability to render an object’s colours ‘naturally’ compared with a familiar reference source (either daylight or an incandescent light).

The colour-rendering index is a set of ratios that provides a quantitative indication of a light source’s ability to reproduce the colours of various objects faithfully compared with an ideal or natural light source, such as the D65 ‘daylight’ illuminant (used to express the spectrum of light corresponding to open air daylight under clear skies in Western or Northern Europe).

The index comprises a set of 14 ratios as defined by CIE (15 with Asian skin tone also included as defined by JIS) that represent agreed test colour samples. The Ra value is the average of the first eight sample ratios, which provides an approximation of the light source colour quality as a single value.

Different light sources can have widely different CRI values. Under the worst types of light source for colour rendering, such as the low-pressure sodium lamp, most colours are not rendered well at all, resulting in a poor illumination quality. Gaps in the spectra of a light source will lead to poor colour rendering. For example, shining orange low-pressure sodium light on a green apple will make the apple look grey in appearance because the light source doesn’t provide any green light for the apple to reflect. Thus, the importance of CRI is obvious to ensure that the objects or people under illumination appear in their full range of colours.

But CRI is a blunt instrument – it’s a single value used to sum up how a light source renders a whole range of colours. There is much debate about the use of the CRI scheme for new light sources such as LEDs. New schemes such as the colour quality scales and colour gamut are being explored to represent better the wider range of light sources available today. Lighting is, however, a complex set of applications. High-CRI lighting that evenly spreads the colour rendition may not be the best light source for every application, and accenting brighter saturated colours might be a better approach.

A retail example

An interesting example of this would be the vivid emphasis of certain colours in a retail environment – making red meat look more red or green apples stand out. In this case, the tuning of the lighting characteristics to ensure high contrast in the red or green colour may createa reduction in the measured CRI value of the light source.

This is because the CRI value is determined by how closely the light source renders colour compared to a daylight source.

In this retail example, a lower CRI but plenty of red or green in the spectrum would be desirable. But when the same light source is used for general lighting – which demands good rendering of a wide range of colours – then this may result in the lighting creating an unnatural appearance such as the accentuation of skin tone or skin blemishes
on people.

Lighting is a complex area and the spectrum of the light source as well as the overall light level in which it is being viewed requires great expertise to ensure good illumination. Thus, the value of the UK’s lighting design community should not be underestimated as we enter the era of solid-state lighting. Light sources of all shapes, sizes and spectral content can be created by many companies with varying experience of lighting. Having light sources appropriately tested in accredited laboratories is also important to ensure that the results accurately represent how they will perform.

Gareth Jones is the CEO of LUX-TSI laboratories, based in the UK and Malaysia.