Predictive Modeling for Public Policy Design: The Impact of Artificial Lights at Night (ALAN) on Bird Strikes

Abstract

Abstract Bird strikes present a growing concern to wildlife preservation efforts and pose safety and economic challenges in aviation and urban design that amalgamates a sociotechnical system. Our review of recent evidence suggests that artificial light at night (ALAN) is a contributing factor to bird strikes, especially during migration seasons, when birds are more likely to be disoriented by bright urban lighting. In this paper, we explore the relationship between light pollution and bird strikes in aviation, with a focus on the application of predictive modeling to inform public policy. Central to our exploration lies is the foundational paradigm of Cyber-Physical-Social Systems (CPSS) that underscores the interaction between the cyber, physical, and social spaces, ensuring effective model-based system design. Multiple criteria are used to perform exploratory data analysis and train a predictive model, including the FAA wildlife strike database, a light pollution GIS map, and BirdCast migration forecasts. Through exploratory data analysis and predictive modeling techniques, we identify critical correlations between light pollution levels and bird strike incidents. These findings are used to reveal significant trends and offer a model capable of predicting bird strike occurrences with implications for urban planning, lighting design, and aviation safety. We emphasize the potential of these predictive models in informing public policy decisions, aiming to mitigate bird strikes while considering ecological and industrial factors through model-based systems design. Moreover, we acknowledge our method’s limitations and provide recommendations for future research to refine the model’s accuracy and applicability in a policy-making context.