Using high-resolution nighttime satellite data to quantify light emissions in Hong Kong

Artificial light at night (ALAN) is a form of anthropogenic pollution that alters the natural and urban environment, affecting the rhythms of flora, fauna and human beings. To mitigate light pollution, it is critical to analyze the sources of ALAN. In this study, we analyze the sources of ALAN using a high-resolution, 1 m spatial resolution satellite image. Through reconstructing individual buildings as seen in the image using a building dataset with height considered, we examined the percentages of ALAN sources across 67 neighborhoods in downtown Hong Kong on both sides of Victoria Harbor. Among the three types of light emissions, building light emissions accounted for the majority (41.8%), followed by all other areas that are mainly streets (41.1%) and parks (17.1%). Despite their limited presence, parks contributed disproportionately to light pollution due to the excessive use of high-power stadium lighting for outdoor sports activities such as basketball, football, and horse racing. Building light emissions primarily came from LED advertisement signboards on office buildings and shopping malls in commercial districts. Nearby residential buildings had their walls illuminated by the brightened urban environment, which, like mirrors, also contributed to light pollution. The percentage of lit buildings is a strong indicator of light pollution severity in a neighborhood, with a R-squared value of 0.76 in relation to total light emissions per area. This study indicates that reducing unnecessary illuminated advertising signboards on commercial buildings provides a highly cost-effective and readily implementable approach to mitigating light pollution.
© 2026 Elsevier B.V.