In recent years, the growing concern over light pollution has sparked significant scientific inquiry into its environmental and societal impacts. Light pollution refers to the excessive or misdirected artificial lighting that brightens the night sky, thereby interfering with natural light-dark cycles. This phenomenon has escalated into a pervasive global environmental challenge that disrupts ecosystems, adversely affects human health, and compromises astronomical observations while sidelining energy conservation efforts. The Department of Physics at The University of Hong Kong (HKU) has recently contributed groundbreaking insights into this issue through a comprehensive study focusing on urban lighting patterns in Hong Kong, revealing how just a handful of decorative lighting installations disproportionately contribute to the city’s night sky brightness.
Hong Kong, known for its vibrant urban landscape, offers a unique case study to examine light pollution because of its dense metropolitan environment interspersed with natural coastal areas. Utilizing innovative multi-source data collection methods, HKU researchers have established that the spatial distribution of light pollution is far from uniform. Instead, it clusters around specific urban districts and is heavily influenced by certain types of lighting fixtures, notably decorative lighting on commercial buildings. Detailed analysis shows that only around 120 buildings with ornamental lights are responsible for approximately half of the light intensity recorded above Victoria Harbour, demonstrating the outsized influence of targeted lighting on overall light pollution levels.
The research team leveraged the annual Earth Hour lights-out initiative, an event that encourages the temporary shutdown of non-essential lighting to raise environmental awareness. By analyzing over a decade of data from HKU’s night sky monitoring stations, investigators observed a dramatic reduction, up to 50%, in sky brightness during the one-hour switch-off period. This “natural experiment” provided a rare opportunity to isolate the effects of these decorative lights on urban light pollution. The researchers also employed a variety of supplementary data sources, including tram-mounted cameras, public smartphone footage, and surveillance images, to identify and characterize the major sources of extraneous light during these events.
.adsslot_aoAdSJRhcK{width:728px !important;height:90px !important;}
@media(max-width:1199px){ .adsslot_aoAdSJRhcK{width:468px !important;height:60px !important;}
}
@media(max-width:767px){ .adsslot_aoAdSJRhcK{width:320px !important;height:50px !important;}
}
ADVERTISEMENT
The investigative approach went beyond traditional sky brightness measures by integrating omnidirectional imaging and extensive field observations, capturing dynamic changes in lighting conditions throughout the cityscape. The tram video recordings were particularly instrumental in highlighting the dramatic contrast in lighting conditions between Earth Hour and typical nights, underlining how few lighting sources exert a dominant effect. This heightened focus on decorative lighting fixtures attributes a central role to billboards and floodlights, which flood the urban environment with unwanted luminosity, significantly contributing to light trespass and skyglow.
Another critical dimension of this study involved the processing of ultra-high-resolution nighttime satellite imagery, captured by the Jilin-1 satellite constellation, which provides spatial resolutions at approximately one meter. This level of granularity allowed the team to perform a finely tuned geographic information system (GIS) analysis, matching light emissions with specific buildings and urban structures. The findings underscored a striking imbalance in light distribution: just five small districts — Central, Happy Valley, Tsim Sha Tsui West, East Tsim Sha Tsui & King’s Park, and Causeway Bay — collectively emit over 50% of the external light detected across both sides of the Victoria Harbour.
The satellite-based analysis also uncovered a nuanced breakdown of light source contributions across different urban categories. Buildings themselves accounted for 42% of detectable light emissions, while parks and sports facilities contributed 17%, with streets, highways, and miscellaneous sources making up the remaining 41%. An unexpected revelation was the significant influence of residential buildings, an area often overlooked in light pollution research. This suggests a complex social dimension to light nuisance, inviting policymakers to reconsider current strategies aimed at managing illumination in densely populated residential zones, where light trespass often infringes upon personal comfort and health.
Beyond mapping and monitoring, the research team exhibited ingenuity by incorporating artificial intelligence techniques to further scrutinize light pollution. Utilizing deep learning algorithms for object detection and image segmentation, they innovated an External Lighting Index, capable of automatically classifying various light sources from publicly submitted smartphone photographs. This AI-driven tool not only enhances research accuracy but also democratizes data collection by involving the public, empowering citizens to contribute valuable environmental data. Through a dedicated web platform, users submit nightly images, building an interactive citizen science protocol that amplifies communal awareness and participation in light pollution mitigation efforts.
The collective findings of the HKU team vividly illustrate the disproportionate contribution of a small subset of decorative and commercial lighting to the urban night glow, suggesting that targeted regulatory actions could yield significant improvements in sky quality. Policy recommendations emerging from the study advocate for more stringent controls on non-essential commercial lighting, retrofitting of building illumination standards, and the development of comprehensive light pollution metrics that can guide sustainable urban lighting design. Implementation of these strategies would not only conserve energy but enhance urban environmental quality without sacrificing the functional and aesthetic benefits of nocturnal cityscapes.
This research owes its success to extensive funding and collaboration, principally supported by the Environment and Conservation Fund of the Hong Kong Special Administrative Region and internal grants from HKU’s Faculty RAE Improvement Fund. These resources enabled the establishment of an extensive night sky monitoring network, acquisition of cutting-edge satellite imagery, and detailed on-the-ground surveys during key light reduction events. The interdisciplinary methodology exemplifies how partnerships between government, academia, and public stakeholders can effectively address multifaceted environmental problems like light pollution.
Crucially, the innovative blend of technological applications, ranging from ground-based light quality measurements to satellite remote sensing and artificial intelligence, sets a new benchmark for urban environmental monitoring. Deploying Sky Quality Meter devices strategically located throughout Hong Kong since 2010, the HKU team contributed to an evolving global network of monitoring stations under the Globe at Night initiative. Continuous monitoring at half-minute intervals has provided rich datasets for assessing temporal trends and event-based impacts, such as those linked to Earth Hour, furnishing empirical support for evidence-based policy interventions.
In summary, the research from HKU elucidates the dramatic effects of a limited number of decorative lighting sources on urban light pollution and exposes the underlying geographic concentration of these effects in Hong Kong. By demonstrating that strategic reduction or management of identified light emissions could halve sky brightness, this study heralds a new avenue in urban environmental science that balances vibrant city life with ecological and human health considerations. The integration of multi-modal data collection, citizen science engagement, and AI-driven analytics offers a scalable framework applicable to other metropolitan regions struggling with the consequences of artificial nocturnal illumination.
Subject of Research: Not applicable
Article Title: Natural experiments from Earth Hour reveal urban night sky being drastically lit up by few decorative buildings
News Publication Date: 1-Jul-2025
Web References:
Scientific Reports article: https://www.nature.com/articles/s41598-025-05279-4
IEEE IGARSS 2023 paper: https://ieeexplore.ieee.org/abstract/document/10283198
References:
Pun, J.C.S. et al., “Natural experiments from Earth Hour reveal urban night sky being drastically lit up by few decorative buildings,” Scientific Reports, 2025.
Pun, J.C.S. et al., “Using High-Resolution Nighttime Remote Sensing Data to Identify Light Sources in Hong Kong,” IGARSS 2023 – IEEE International Geoscience and Remote Sensing Symposium, 2023.
Image Credits: Hong Kong Light Pollution Research Project Team, The University of Hong Kong
Keywords
Physical sciences, Physics, Light Pollution, Urban Environment, Remote Sensing, Artificial Intelligence, Environmental Monitoring
Tags: decorative lighting contributions to light pollutionenvironmental impacts of light pollutionHKU research on urban lightinghuman health and light pollutionimpact of artificial lighting on ecosystemsinnovative data collection methods in researchlight pollution and energy conservationlight pollution in Hong Kongnight sky brightness in urban areasspatial distribution of light pollutionurban landscape and light pollutionurban light pollution effects
