Limitations of a Historically Important but Outdated Design Approach
The L20 method represents one of the earliest systematic approaches developed for tunnel entrance lighting design. For decades, it was widely adopted because it offered a relatively simple and practical way to estimate the required lighting level at tunnel entrances. The fundamental premise of the method is straightforward: the brighter the environment outside the tunnel, the higher the luminance required inside the entrance zone.
L20 is defined as the average luminance of the environment observed by a driver within a 20-degree cone of vision, measured from the stopping distance in front of the tunnel portal. When introduced, this method provided a significant improvement over purely empirical design practices and helped standardize tunnel lighting calculations across many countries.
However, extensive research and operational experience have revealed that the L20 method suffers from fundamental conceptual limitations that prevent it from accurately representing real driving conditions at tunnel entrances.
The most critical limitation of L20 is that it considers only physical exterior luminance. It assumes that the driver’s visual performance is directly related to the brightness of the surrounding environment. In reality, the driver’s ability to perceive objects depends not only on brightness but on perceived contrast, which is strongly affected by veiling luminance.
Veiling luminance arises from multiple sources:
- Light scattered in the atmosphere
- Reflections and diffusion at the vehicle windshield
- Reflections from the vehicle interior
- Intraocular light scattering
None of these effects are considered in the L20 method. As a result, L20 can significantly underestimate the lighting required to ensure safe visibility, particularly under conditions of high exterior brightness.
The International Commission on Illumination explicitly acknowledges these limitations in CIE 88 – Guide for the Lighting of Road Tunnels and Underpasses. The guide states that exterior luminance alone is not sufficient to assess visibility conditions at tunnel entrances and that additional perceptual factors must be considered.
Another important drawback of L20 is its poor adaptability to complex environments. Tunnel entrances located in mountainous regions, urban canyons, snow-covered landscapes, or areas with highly reflective surfaces can produce dramatically different visual conditions even when L20 values appear similar. The method lacks the ability to distinguish between these scenarios, leading to inconsistent and sometimes unsafe design outcomes.
In summary, while L20 played a valuable historical role in the evolution of tunnel lighting design, it does not adequately represent the driver’s real visual experience. Modern tunnel lighting design requires a more comprehensive, perception-based approach that goes beyond exterior brightness alone.