Assessment of melanopsin-based quantities: Comparison of selected design tools and validation against on-field measurements

Melanopsin-based quantities (circadian stimulus and melanopic illuminance) depend on the spectral irradiance at the eye level, in turn resulting from the light sources' spectral power distribution and the interactions between light and materials. Software used in design practice (as DIALux) do not consider light and materials' spectral characteristics. Consequently, circadian stimulus and melanopic illuminance are calculated based on the illuminance at the eye and the sources' spectrum. Although software including spectral interactions (as ALFA - Adaptive Lighting for Alertness) exist, it is not clear how much the accuracy of various modelling approaches affects melanopsin-based quantities assessment. This research aims at investigating the applicability of the popular lighting design software for evaluating non-visual effects. For this purpose, two different approaches, based on the use of ALFA (method A) and DIALux (method B) are described, validated against on-field measurements and compared. Spectral irradiance is measured at the work plane and at the eye in a test room where the walls' finishes and the luminaires’ spectrum are changed, obtaining 21 scenarios. When simulating the space with the two methods, acceptable results are obtained for horizontal illuminance (percentage errors within the range ±10%), whereas for the vertical plane, errors depend on the software and the lighting scenario. Overall, method A performs better than B, but it gives percentage errors trespassing ±10% in 38.10% and 52.38% of the scenes in estimating circadian stimulus and melanopic illuminance respectively. This research informs ALFA and DIALux users on the reliability of both simulation tools in melanopsin-based quantities evaluation.