Are standard r-tables still representative of road surface photometric characteristics in France?

Road lighting is intended to improve the comfort and safety of road users at night by increasing visibility levels. This purpose is best achieved by ensuring certain luminance levels and uniformities on the pavement. The luminance distribution depends on the photometric properties of both the luminaries and the pavement. Hence, the reflection properties of the pavement must be taken into account when designing a road lighting installation (CIE 66-1984; CIE 140-2000). For road lighting purposes, pavement reflection properties are specified by means of the reduced luminance coefficient r, given in a two dimensional table for a number of standardized combinations of the lighting angle g and the reflection angle b. This so called r-table is difficult to determine, so standard tables have been introduced to facilitate lighting calculations, based on a classification of dry road surfaces with respect to their degree of lightness Q0 and their degree of specularity S1. The four-class system R is used in France; other countries may use different systems, but even the most recent dates from the late 70s'. Thus the values contained in the standard tables are characteristic of materials that were used decades ago. Practices in road construction have changed since that time, and new pavement materials have been used. So whether the standard data still applies to current road surfaces is a relevant question (Brusque et al, 1997; Fotios et al, 2006). To investigate on that question, the LCPC organized a series of in situ measurements to gather up-to-date data on the reflection properties of typical pavement materials, and how these properties evolve over time. Samples were taken periodically over a period of up to three years, on road sections which had been re-surfaced for that purpose. Laboratory measurements were performed by the LRPC of Clermont-Ferrand, using a CIE-compliant goniophotometer (CIE 66-1984). The resulting photometric data was analysed in terms of lightness and specularity to estimate the influence of several factors: material type, aggregate size and colour, and age. Very thin bituminous concretes are most common in France. They have high specularity values (R4) at the young age, which rapidly decrease in the first few months. After three years, they end up fairly or slightly specular (R3 or R2), and even diffuse for those with white aggregates (R1). Porous asphalts evolve in a similar way (from R1 to R3 or R2), only they stabilize faster, and they are much darker. Surface dressings on the other hand are quite diffuse at the start (R1), but their specularity increases over time. They end up more or less specular (R3 or R2), depending on the size of the aggregates, after about two years. Measured r-tables were also graphically compared to the standard tables of the corresponding class. Two types of discrepancies were observed. In some cases, particularly for diffuse materials, the tables only differed by their lightness value. But in other cases, the shape of the r-body itself differed, even after scaling the standard r-table to match lightness values. As a conclusion, there is no doubt that the reflection properties of current pavement materials in France have diverged from the standard tables. The economic impact of this discrepancy has been demonstrated by the CERTU (Chain et al, 2007)