Towards an absolute light pollution indicator.

The growing concern about the negative impact of artificial light at night on biodiversity and human health increases the need of defining a general indicator that could be used for characterizing light pollution as well as performing both spatial and temporal comparisons. In this paper, we show that the traditional indicators based on direct numerical measurements of sky brightness suffer from significant limitation due to calibration bias and lack of reproducibility. Furthermore, these measures are most often performed in periods of clear sky.