A systematic light pollution modelling bias in present night sky brightness predictions

Night sky brightness (NSB) measurements inform both the selection of new ground-based optical and near-infrared astronomical observatory sites and the management of night-time conditions near existing facilities. NSB modelling supports site characterization in cases where obtaining in situ measurements is difficult and provides additional information about the photic environment. Models of NSB conventionally assume Mie scattering, but we show that predominantly at low-altitude sites contaminated by artificial light, this assumption may result in significant errors up to a factor of ~2.5. This effect introduces systematic bias to models of anthropogenic skyglow, but more realistic modelling approaches tend to be computationally intensive. We demonstrate the significance of these effects by simulating NSB while accounting for increasingly realistic populations of aerosol components. The use of approximate scattering phase functions reduces computing time while yielding results that reasonably match field observations. This approach can speed up models used to predict NSB while increasing their accuracy.